| gm(1) | General Commands Manual | gm(1) |
gm - command-line utility to create, edit, compare, convert, or display images
gm animate [ options ... ] file [ [ options ... ] file ... ]
gm batch [ options ... ] [ script ]
gm benchmark [ options ... ] subcommand
gm compare [ options ... ] reference-image [ options ... ] compare-image [ options ... ]
gm composite [ options ... ] change-image base-image [ mask-image ] output-image
gm conjure [ options ] script.msl [ [ options ] script.msl ]
gm convert [ [ options ... ] [ input-file ... ] [ options ... ] ] output-file
gm display [ options ... ] file ... [ [options ... ]file ... ]
gm identify file [ file ... ]
gm import [ options ... ] file
gm mogrify [ options ... ] file ...
gm montage [ options ... ] file [ [ options ... ] file ... ] output-file
gm time subcommand
gm version
GraphicsMagick's gm provides a suite of utilities for creating, comparing, converting, editing, and displaying images. All of the utilities are provided as sub-commands of a single gm executable. The gm executable returns the exit code 0 to indicate success, or 1 to indicate failure:
animate displays an animation (e.g. a GIF file) on any workstation display running an X server.
batch executes an arbitary number of the utility commands (e.g. convert) in the form of a simple linear batch script in order to improve execution efficiency, and/or to allow use as a subordinate co-process under the control of an arbitrary script or program.
benchmark executes one of the other utility commands (e.g. convert) for a specified number of iterations, or execution time, and reports execution time and other profiling information such as CPU utilization. Benchmark provides various operating modes including executing the command with a varying number of threads, and alternate reporting formats such as comma-separated value (CSV).
compare compares two images and reports difference statistics according to specified metrics and/or outputs an image with a visual representation of the differences. It may also be used to test if images are similar within a particular range and specified metric, returning a truth value to the executing environment.
composite composites images (blends or merges images together) to create new images.
conjure interprets and executes scripts in the Magick Scripting Language (MSL).
convert converts an input file using one image format to an output file with the same or differing image format while applying an arbitrary number of image transformations.
display is a machine architecture independent image processing and display facility. It can display an image on any workstation display running an X server.
identify describes the format and characteristics of one or more image files. It will also report if an image is incomplete or corrupt.
import reads an image from any visible window on an X server and outputs it as an image file. You can capture a single window, the entire screen, or any rectangular portion of the screen.
mogrify transforms an image or a sequence of images. These transforms include image scaling, image rotation, color reduction, and others. The transmogrified image overwrites the original image.
montage creates a composite by combining several separate images. The images are tiled on the composite image with the name of the image optionally appearing just below the individual tile.
time executes a subcommand and reports the user, system, and total execution time consumed.
version reports the GraphicsMagick release version, maximum sample-depth, copyright notice, supported features, and the options used while building the software.
The GraphicsMagick utilities recognize the following image formats:
Name Mode Description
o 3FR r-- Hasselblad Photo RAW
o 8BIM rw- Photoshop resource format
o 8BIMTEXT rw- Photoshop resource text format
o 8BIMWTEXT rw- Photoshop resource wide text format
o APP1 rw- Raw application information
o APP1JPEG rw- Raw JPEG binary data
o ART r-- PF1: 1st Publisher
o ARW r-- Sony Alpha DSLR RAW
o AVS rw+ AVS X image
o BIE rw- Joint Bi-level Image experts Group
interchange format
o BMP rw+ Microsoft Windows bitmap image
o BMP2 -w- Microsoft Windows bitmap image v2
o BMP3 -w- Microsoft Windows bitmap image v3
o CACHE --- Magick Persistent Cache image format
o CALS rw- Continuous Acquisition and Life-cycle
Support Type 1 image
o CAPTION r-- Caption (requires separate size info)
o CIN rw- Kodak Cineon Format
o CMYK rw- Raw cyan, magenta, yellow, and black
samples (8 or 16 bits, depending on
the image depth)
o CMYKA rw- Raw cyan, magenta, yellow, black, and
matte samples (8 or 16 bits, depending
on the image depth)
o CR2 r-- Canon Photo RAW
o CRW r-- Canon Photo RAW
o CUR r-- Microsoft Cursor Icon
o CUT r-- DR Halo
o DCM r-- Digital Imaging and Communications in
Medicine image
o DCR r-- Kodak Photo RAW
o DCX rw+ ZSoft IBM PC multi-page Paintbrush
o DNG r-- Adobe Digital Negative
o DPS r-- Display PostScript Interpreter
o DPX rw- Digital Moving Picture Exchange
o EPDF rw- Encapsulated Portable Document Format
o EPI rw- Adobe Encapsulated PostScript
Interchange format
o EPS rw- Adobe Encapsulated PostScript
o EPS2 -w- Adobe Level II Encapsulated PostScript
o EPS3 -w- Adobe Level III Encapsulated PostScript
o EPSF rw- Adobe Encapsulated PostScript
o EPSI rw- Adobe Encapsulated PostScript
Interchange format
o EPT rw- Adobe Encapsulated PostScript with MS-DOS
TIFF preview
o EPT2 rw- Adobe Level II Encapsulated PostScript
with MS-DOS TIFF preview
o EPT3 rw- Adobe Level III Encapsulated PostScript
with MS-DOS TIFF preview
o EXIF rw- Exif digital camera binary data
o FAX rw+ Group 3 FAX (Not TIFF Group3 FAX!)
o FITS rw- Flexible Image Transport System
o FRACTAL r-- Plasma fractal image
o FPX rw- FlashPix Format
o GIF rw+ CompuServe graphics interchange format
o GIF87 rw- CompuServe graphics interchange format
(version 87a)
o GRADIENT r-- Gradual passing from one shade to
another
o GRAY rw+ Raw gray samples (8/16/32 bits,
depending on the image depth)
o HISTOGRAM -w- Histogram of the image
o HRZ r-- HRZ: Slow scan TV
o HTML -w- Hypertext Markup Language and a
client-side image map
o ICB rw+ Truevision Targa image
o ICC rw- ICC Color Profile
o ICM rw- ICC Color Profile
o ICO r-- Microsoft icon
o ICON r-- Microsoft icon
o IDENTITY r-- Hald CLUT identity image
o IMAGE r-- GraphicsMagick Embedded Image
o INFO -w+ Image descriptive information and
statistics
o IPTC rw- IPTC Newsphoto
o IPTCTEXT rw- IPTC Newsphoto text format
o IPTCWTEXT rw- IPTC Newsphoto wide text format
o JBG rw+ Joint Bi-level Image experts Group
interchange format
o JBIG rw+ Joint Bi-level Image experts Group
interchange format
o JNG rw- JPEG Network Graphics
o JP2 rw- JPEG-2000 JP2 File Format Syntax
o JPC rw- JPEG-2000 Code Stream Syntax
o JPEG rw- Joint Photographic Experts Group
JFIF format
o JPG rw- Joint Photographic Experts Group
JFIF format
o K25 r-- Kodak Photo RAW
o KDC r-- Kodak Photo RAW
o LABEL r-- Text image format
o M2V rw+ MPEG-2 Video Stream
o MAP rw- Colormap intensities and indices
o MAT r-- MATLAB image format
o MATTE -w+ MATTE format
o MIFF rw+ Magick Image File Format
o MNG rw+ Multiple-image Network Graphics
o MONO rw- Bi-level bitmap in least-significant-
-byte-first order
o MPC rw+ Magick Persistent Cache image format
o MPEG rw+ MPEG-1 Video Stream
o MPG rw+ MPEG-1 Video Stream
o MRW r-- Minolta Photo Raw
o MSL r-- Magick Scripting Language
o MTV rw+ MTV Raytracing image format
o MVG rw- Magick Vector Graphics
o NEF r-- Nikon Electronic Format
o NULL r-- Constant image of uniform color
o OTB rw- On-the-air bitmap
o P7 rw+ Xv thumbnail format
o PAL rw- 16bit/pixel interleaved YUV
o PALM rw- Palm Pixmap
o PBM rw+ Portable bitmap format (black and white)
o PCD rw- Photo CD
o PCDS rw- Photo CD
o PCL -w- Page Control Language
o PCT rw- Apple Macintosh QuickDraw/PICT
o PCX rw- ZSoft IBM PC Paintbrush
o PDB rw+ Palm Database ImageViewer Format
o PDF rw+ Portable Document Format
o PEF r-- Pentax Electronic File
o PFA r-- TrueType font
o PFB r-- TrueType font
o PGM rw+ Portable graymap format (gray scale)
o PGX r-- JPEG-2000 VM Format
o PICON rw- Personal Icon
o PICT rw- Apple Macintosh QuickDraw/PICT
o PIX r-- Alias/Wavefront RLE image format
o PLASMA r-- Plasma fractal image
o PNG rw- Portable Network Graphics
o PNG24 rw- Portable Network Graphics, 24 bit RGB
opaque only
o PNG32 rw- Portable Network Graphics, 32 bit RGBA
semitransparency OK
o PNG8 rw- Portable Network Graphics, 8-bit
indexed, binary transparency only
o PNM rw+ Portable anymap
o PPM rw+ Portable pixmap format (color)
o PREVIEW -w- Show a preview an image enhancement,
effect, or f/x
o PS rw+ Adobe PostScript
o PS2 -w+ Adobe Level II PostScript
o PS3 -w+ Adobe Level III PostScript
o PSD rw- Adobe Photoshop bitmap
o PTIF rw- Pyramid encoded TIFF
o PWP r-- Seattle Film Works
o RAF r-- Fuji Photo RAW
o RAS rw+ SUN Rasterfile
o RGB rw+ Raw red, green, and blue samples
o RGBA rw+ Raw red, green, blue, and matte samples
o RLA r-- Alias/Wavefront image
o RLE r-- Utah Run length encoded image
o SCT r-- Scitex HandShake
o SFW r-- Seattle Film Works
o SGI rw+ Irix RGB image
o SHTML -w- Hypertext Markup Language and a
client-side image map
o STEGANO r-- Steganographic image
o SUN rw+ SUN Rasterfile
o SVG rw+ Scalable Vector Gaphics
o TEXT rw+ Raw text
o TGA rw+ Truevision Targa image
o TIFF rw+ Tagged Image File Format
o TILE r-- Tile image with a texture
o TIM r-- PSX TIM
o TOPOL r-- TOPOL X Image
o TTF r-- TrueType font
o TXT rw+ Raw text
o UIL -w- X-Motif UIL table
o UYVY rw- 16bit/pixel interleaved YUV
o VDA rw+ Truevision Targa image
o VICAR rw- VICAR rasterfile format
o VID rw+ Visual Image Directory
o VIFF rw+ Khoros Visualization image
o VST rw+ Truevision Targa image
o WBMP rw- Wireless Bitmap (level 0) image
o WMF r-- Windows Metafile
o WPG r-- Word Perfect Graphics
o X rw- X Image
o X3F r-- Foveon X3 (Sigma/Polaroid) RAW
o XBM rw- X Windows system bitmap (black
and white)
o XC r-- Constant image uniform color
o XCF r-- GIMP image
o XMP rw- Adobe XML metadata
o XPM rw- X Windows system pixmap (color)
o XV rw+ Khoros Visualization image
o XWD rw- X Windows system window dump (color)
o YUV rw- CCIR 601 4:1:1 or 4:2:2 (8-bit only)
Modes:
r Read
w Write
+ Multi-image
Support for some of these formats require additional programs or libraries. See README in the source package for where to find optional additional software.
Note, a format delineated with + means that if more than one image is specified, frames are combined into a single multi-image file. Use +adjoin if you want a single image produced for each frame.
Your installation might not support all of the formats in the list. To get an accurate listing of the formats supported by your particular configuration, run "gm convert -list format".
Raw images are expected to have one byte per pixel unless gm is compiled in 16-bit quantum mode or in 32-bit quantum mode. Here, the raw data is expected to be stored two or four bytes per pixel, respectively, in most-significant-byte-first order. For example, you can tell if gm was compiled in 16-bit mode by typing "gm version" without any options, and looking for "Q:16" in the first line of output.
By default, the image format is determined by its magic number, i.e., the first few bytes of the file. To specify a particular image format, precede the filename with an image format name and a colon (i.e.ps:image) or specify the image type as the filename suffix (i.e.image.ps). The magic number takes precedence over the filename suffix and the prefix takes precedence over the magic number and the suffix in input files. When a file is read, its magic number is stored in the "image->magick" string. In output files, the prefix takes precedence over the filename suffix, and the filename suffix takes precedence over the "image->magick" string.
To read the "built-in" formats (GRANITE, H, LOGO, NETSCAPE, PLASMA, and ROSE) use a prefix (including the colon) without a filename or suffix. To read the XC format, follow the colon with a color specification. To read the CAPTION format, follow the colon with a text string or with a filename prefixed with the at symbol (@).
When you specify X as your image type, the filename has special meaning. It specifies an X window by id, name, or root. If no filename is specified, the window is selected by clicking the mouse in the desired window.
Specify input_file as - for standard input, output_file as - for standard output. If input_file has the extension .Z or .gz, the file is uncompressed with uncompress or gunzip respectively. If output_file has the extension .Z or .gz, the file is compressed using with compress or gzip respectively.
Use an optional index enclosed in brackets after an input file name to specify a desired subimage of a multi-resolution image format like Photo CD (e.g. "img0001.pcd[4]") or a range for MPEG images (e.g. "video.mpg[50-75]"). A subimage specification can be disjoint (e.g. "image.tiff[2,7,4]"). For raw images, specify a subimage with a geometry (e.g. -size 640x512 "image.rgb[320x256+50+50]"). Surround the image name with quotation marks to prevent your shell from interpreting the square brackets. Single images are written with the filename you specify. However, multi-part images (e.g., a multi-page PostScript document with +adjoin specified) may be written with the scene number included as part of the filename. In order to include the scene number in the filename, it is necessary to include a printf-style %d format specification in the file name and use the +adjoin option. For example,
image%02d.miff
writes files image00.miff, image01.miff, etc. Only a single specification is allowed within an output filename. If more than one specification is present, it will be ignored. It is best to embed the scene number in the base part of the file name, not in the extension, because the extension will not be a recognizeable image type.
When running a commandline utility, you can prepend an at sign @ to a filename to read a list of image filenames from that file. This is convenient in the event you have too many image filenames to fit on the command line.
Options are processed in command line order. Any option you specify on the command line remains in effect for the set of images that follows, until the set is terminated by the appearance of any option or -noop. Some options only affect the decoding of images and others only the encoding. The latter can appear after the final group of input images.
This is a combined list of the command-line options used by the GraphicsMagick utilities (animate, compare, composite, convert, display, identify, import, mogrify and montage).
In this document, angle brackets ("<>") enclose variables and curly brackets ("{}") enclose optional parameters. For example, "-fuzz <distance>{%}" means you can use the option "-fuzz 10" or "-fuzz 2%".
By default, all images of an image sequence are stored in the same file. However, some formats (e.g. JPEG) do not support storing more than one image per file and only the first frame in an image sequence will be saved unless the result is saved to separate files. Use +adjoin to force saving multiple frames to multiple numbered files. If +adjoin is used, then the output filename must include a printf style formatting specification for the numeric part of the filename. For example,
image%02d.miff
This option provides a transform matrix {sx,rx,ry,sy,tx,ty} for use by subsequent -draw or -transform options.
By default antialiasing algorithms are used when drawing objects (e.g. lines) or rendering vector formats (e.g. WMF and Postscript). Use +antialias to disable use of antialiasing algorithms. Reasons to disable antialiasing include avoiding increasing colors in the image, or improving rendering speed.
This option creates a single image where the images in the original set are stacked top-to-bottom. If they are not of the same width, any narrow images will be expanded to fit using the background color. Use +append to stack images left-to-right. The set of images is terminated by the appearance of any option. If the -append option appears after all of the input images, all images are appended.
Applies ("bakes in") the ASC CDL, which is a format for the exchange of basic primary color grading information between equipment and software from different manufacturers. The format defines the math for three functions: slope, offset and power. Each function uses a number for the red, green, and blue color channels for a total of nine numbers comprising a single color decision. The tenth number (optional) is for chromiance (saturation) as specified by ASC CDL 1.2.
The argument string is comma delimited and is in the following form (but without invervening spaces or line breaks)
redslope,redoffset,redpower:
greenslope,greenoffset,greenpower:
blueslope,blueoffset,bluepower:
saturation
with the unity (no change) specification being:
"1.0,0.0,1.0:1.0,0.0,1.0:1.0,0.0,1.0:1.0"
Use this option to supply a password for decrypting an image or an image sequence, if it is being read from a format such as PDF that supports encryption. Encrypting images being written is not supported.
Adjusts the image orienation so that it is suitable for viewing. Uses the orientation tag obtained from the image file or as supplied by the -orient option.
The set of images is terminated by the appearance of any option. If the -average option appears after all of the input images, all images are averaged.
This backdrop covers the entire workstation screen and is useful for hiding other X window activity while viewing the image. The color of the backdrop is specified as the foreground color (X11 default is black). Refer to "X Resources", below, for details.
The color is specified using the format described under the -fill option.
Use -black-threshold to set pixels with values below the specified threshold to minimum value (black). If only one value is supplied, or the red, green, and blue values are identical, then intensity thresholding is used. If the color threshold values are not identical then channel-based thresholding is used, and color distortion will occur. Specify a negative value (e.g. -1) if you want a channel to be ignored but you do want to threshold a channel later in the list. If a percent (%) symbol is appended, then the values are treated as a percentage of maximum range.
Blur with the given radius and standard deviation (sigma).
See -geometry for details about the geometry specification.
The color is specified using the format described under the -fill option.
The color is specified using the format described under the -fill option.
See -draw for further details.
Choose from: Red, Green, Blue, Opacity, Matte, Cyan, Magenta, Yellow, Black, or Gray.
Use this option to extract a particular channel from the image. Opacity, for example, is useful for extracting the opacity values from an image.
Width and height give the number of columns and rows to remove, and x and y are offsets that give the location of the leftmost column and topmost row to remove.
The x offset normally specifies the leftmost column to remove. If the -gravity option is present with NorthEast, East, or SouthEast gravity, it gives the distance leftward from the right edge of the image to the rightmost column to remove. Similarly, the y offset normally specifies the topmost row to remove, but if the -gravity option is present with SouthWest, South, or SouthEast gravity, it specifies the distance upward from the bottom edge of the image to the bottom row to remove.
The -chop option removes entire rows and columns, and moves the remaining corner blocks leftward and upward to close the gaps.
If a clipping path is present, it will be applied to subsequent operations.
For example, if you type the following command:
gm convert -clip -negate cockatoo.tif negated.tif
only the pixels within the clipping path are negated.
The -clip feature requires the XML library. If the XML library is not present, the option is ignored.
Each image N in the sequence after Image 0 is replaced with the image created by flattening images 0 through N.
The set of images is terminated by the appearance of any option. If the -coalesce option appears after all of the input images, all images are coalesced.
Specify the amount of colorization as a percentage. You can apply separate colorization values to the red, green, and blue channels of the image with a colorization value list delimited with slashes (e.g. 0/0/50).
The -colorize option may be used in conjunction with -modulate to produce a nice sepia toned image like:
gm convert input.ppm -modulate 115,0,100 \
-colorize 7,21,50 output.ppm.
Choose between shared or private.
This option only applies when the default X server visual is PseudoColor or GRAYScale. Refer to -visual for more details. By default, a shared colormap is allocated. The image shares colors with other X clients. Some image colors could be approximated, therefore your image may look very different than intended. Choose Private and the image colors appear exactly as they are defined. However, other clients may go technicolor when the image colormap is installed.
The actual number of colors in the image may be less than your request, but never more. Note, this is a color reduction option. Images with less unique colors than specified with this option will have any duplicate or unused colors removed. The ordering of an existing color palette may be altered. When converting an image from color to grayscale, convert the image to the gray colorspace before reducing the number of colors since doing so is most efficient. Refer to <a href="quantize.html">quantize for more details.
Note, options -dither, -colorspace, and -treedepth affect the color reduction algorithm.
Choices are: CineonLog, CMYK, GRAY, HSL, HWB, OHTA, RGB, Rec601Luma, Rec709Luma, Rec601YCbCr, Rec709YCbCr, Transparent, XYZ, YCbCr, YIQ, YPbPr, or YUV.
Color reduction, by default, takes place in the RGB color space. Empirical evidence suggests that distances in color spaces such as YUV or YIQ correspond to perceptual color differences more closely than do distances in RGB space. These color spaces may give better results when color reducing an image. Refer to quantize for more details. Two gray colorspaces are supported. The Rec601Luma space is based on the recommendations for legacy NTSC television (ITU-R BT.601-5). The Rec709Luma space is based on the recommendations for HDTV (Rec. ITU-R BT.709-5) and is suitable for use with computer graphics, and for contemporary CRT displays. The GRAY colorspace currently selects the Rec601Luma colorspace by default for backwards compatibly reasons. This default may be re-considered in the future.
Two YCbCr colorspaces are supported. The Rec601YCbCr space is based on the recommendations for legacy NTSC television (ITU-R BT.601-5). The Rec709CbCr space is based on the recommendations for HDTV (Rec. ITU-R BT.709-5) and is suitable for suitable for use with computer graphics, and for contemporary CRT displays. The YCbCr colorspace specification is equivalent toRec601YCbCr.
The Transparent color space behaves uniquely in that it preserves the matte channel of the image if it exists.
The -colors or -monochrome option, or saving to a file format which requires color reduction, is required for this option to take effect.
Use this option to assign a specific comment to the image, when writing to an image format that supports comments. You can include the image filename, type, width, height, or other image attribute by embedding special format characters listed under the -format option. The comment is not drawn on the image, but is embedded in the image datastream via a "Comment" tag or similar mechanism. If you want the comment to be visible on the image itself, use the -draw option instead.
For example,
-comment "%m:%f %wx%h"
produces an image comment of MIFF:bird.miff 512x480 for an image titled bird.miff and whose width is 512 and height is 480.
If the first character of string is @, the image comment is read from a file titled by the remaining characters in the string. Please note that if the string comes from an untrusted source that it should be sanitized before use since otherwise the content of an arbitrary readable file could be incorporated in a comment in the output file (a security risk).
If the -comment option appears multiple times, only the last comment is stored.
In PNG images, the comment is stored in a tEXt or zTXt chunk with the keyword "comment".
The description of composition uses abstract terminology in order to allow the the description to be more clear, while avoiding constant values which are specific to a particular build configuration. Each image pixel is represented by red, green, and blue levels (which are equal for a gray pixel). MaxRGB is the maximum integral value which may be stored in the red, green, or blue channels of the image. Each image pixel may also optionally (if the image matte channel is enabled) have an associated level of opacity (ranging from opaque to transparent), which may be used to determine the influence of the pixel color when compositing the pixel with another image pixel. If the image matte channel is disabled, then all pixels in the image are treated as opaque. The color of an opaque pixel is fully visible while the color of a transparent pixel color is entirely absent (pixel color is ignored).
By definition, raster images have a rectangular shape. All image rows are of equal length, and all image columns have the same number of rows. By treating the opacity channel as a visual "mask" the rectangular image may be given a "shape" by treating the opacity channel as a cookie-cutter for the image. Pixels within the shape are opaque, while pixels outside the shape are transparent. Pixels on the boundary of the shape may be between opaque and transparent in order to provide antialiasing (visually smooth edges). The description of the composition operators use this concept of image "shape" in order to make the description of the operators easier to understand. While it is convenient to describe the operators in terms of "shapes" they are by no means limited to mask-style operations since they are based on continuous floating-point mathematics rather than simple boolean operations.
By default, the Over composite operator is used. The following composite operators are available:
Over
In
Out
Atop
Xor
Plus
Minus
Add
Subtract
Difference
Divide
Multiply
Bumpmap
Copy
CopyRed
CopyGreen
CopyBlue
CopyOpacity
CopyCyan
CopyMagenta
CopyYellow
CopyBlack
The behavior of each operator is described below.
Over
The result will be the union of the two image shapes, with opaque areas of
change-image obscuring base-image in the region of
overlap.
In
The result is simply change-image cut by the shape of
base-image. None of the image data of base-image will be
in the result.
Out
The resulting image is change-image with the shape of
base-image cut out.
Atop
The result is the same shape as base-image, with
change-image obscuring base-image where the image shapes
overlap. Note this differs from over because the portion of
change-image outside base-image's shape does not appear in
the result.
Xor
The result is the image data from both change-image and
base-image that is outside the overlap region. The overlap region
will be blank.
Plus
The result is just the sum of the image data. Output values are cropped to
MaxRGB (no overflow). This operation is independent of the matte
channels.
Minus
The result of change-image - base-image, with underflow
cropped to zero. The matte channel is ignored (set to opaque, full
coverage).
Add
The result of change-image + base-image, with overflow
wrapping around (mod MaxRGB+1).
Subtract
The result of change-image - base-image, with underflow
wrapping around (mod MaxRGB+1). The add and
subtract operators can be used to perform reversible
transformations.
Difference
The result of abs(change-image - base-image). This is useful
for comparing two very similar images.
Divide
The result of change-image / base-image. This is useful for
improving the readability of text on unevenly illuminated photos (by
using a gaussian blurred copy of change-image as base-image).
Multiply
The result of change-image * base-image. This is useful for
the creation of drop-shadows.
Bumpmap
The result base-image shaded by change-image.
Copy
The resulting image is base-image replaced with
change-image. Here the matte information is ignored.
CopyRed
The resulting image is the red channel in base-image replaced with
the red channel in change-image. The other channels are copied
untouched.
CopyGreen
The resulting image is the green channel in base-image replaced
with the green channel in change-image. The other channels are
copied untouched.
CopyBlue
The resulting image is the blue channel in base-image replaced with
the blue channel in change-image. The other channels are copied
untouched.
CopyOpacity
The resulting image is the opacity channel in base-image replaced
with the opacity channel in change-image. The other channels are
copied untouched.
CopyCyan
The resulting image is the cyan channel in base-image replaced with
the cyan channel in change-image. The other channels are copied
untouched. Use of this operator requires that base-image be in CMYK(A)
colorspace.
CopyMagenta
The resulting image is the magenta channel in base-image replaced
with the magenta channel in change-image. The other channels are
copied untouched. Use of this operator requires that base-image be in
CMYK(A) colorspace.
CopyYellow
The resulting image is the yellow channel in base-image replaced
with the yellow channel in change-image. The other channels are
copied untouched. Use of this operator requires that base-image be in
CMYK(A) colorspace.
CopyBlack
The resulting image is the black channel in base-image replaced
with the black channel in change-image. The other channels are
copied untouched. Use of this operator requires that base-image be in
CMYK(A) colorspace. If change-image is not in CMYK space, then the
change-image pixel intensities are used.
Choices are: None, BZip, Fax, Group3, Group4, JPEG, Lossless, LZW, RLE, Zip, LZMA, JPEG2000, JPEG2000, JBIG, JBIG2, WebP, or ZSTD.
Specify +compress to store the binary image in an uncompressed format. The default is the compression type of the specified image file.
"Lossless" refers to lossless JPEG, which is only available if the JPEG library has been patched to support it. Use of lossless JPEG is generally not recommended.
Use the -quality option to set the compression level to be used by the JPEG, JPEG-2000, PNG, MIFF, MPEG, and TIFF encoders. Use the -sampling-factor option to set the sampling factor to be used by the DPX, JPEG, MPEG, and YUV encoders for downsampling the chroma channels.
This option enhances the intensity differences between the lighter and darker elements of the image. Use -contrast to enhance the image or +contrast to reduce the image contrast.
For a more pronounced effect you can repeat the option:
gm convert rose: -contrast -contrast rose_c2.png
The kernel is specified as a comma-separated list of floating point values, ordered left-to right, starting with the top row. The order of the kernel is determined by the square root of the number of entries. Presently only square kernels are supported.
Use this option with -output-directory if the input paths contain subdirectories and it is desired to create similar subdirectories in the output directory. Without this option, mogrify will fail if the required output directory does not exist.
See -geometry for details about the geometry specification.
The width and height give the size of the image that remains after cropping, and x and y are offsets that give the location of the top left corner of the cropped image with respect to the original image. To specify the amount to be removed, use -shave instead.
If the x and y offsets are present, a single image is generated, consisting of the pixels from the cropping region. The offsets specify the location of the upper left corner of the cropping region measured downward and rightward with respect to the upper left corner of the image. If the -gravity option is present with NorthEast, East, or SouthEast gravity, it gives the distance leftward from the right edge of the image to the right edge of the cropping region. Similarly, if the -gravity option is present with SouthWest, South, or SouthEast gravity, the distance is measured upward between the bottom edges.
If the x and y offsets are omitted, a set of tiles of the specified geometry, covering the entire input image, is generated. The rightmost tiles and the bottom tiles are smaller if the specified geometry extends beyond the dimensions of the input image.
Amount defines the number of positions each colormap entry isshifted.
The events parameter specifies which events are to be logged. It can be either None, All, or a comma-separated list consisting of one or more of the following domains: Annotate, Blob, Cache, Coder, Configure, Deprecate, Error, Exception, FatalError, Information, Locale, Option, Render, Resource, TemporaryFile, Transform, User. Warning, or X11, For example, to log cache and blob events, use
gm convert -debug "Cache,Blob" rose: rose.png
The "User" domain is normally empty, but developers can log "User" events in their private copy of GraphicsMagick.
Use the -log option to specify the format for debugging output.
Use +debug to turn off all logging.
An alternative to using -debug is to use the MAGICK_DEBUG environment variable. The allowed values for the MAGICK_DEBUG environment variable are the same as for the -debug option.
This option compares each image with the next in a sequence and returns the maximum bounding region of any pixel differences it discovers. This method can undo a coalesced sequence returned by the -coalesce option, and is useful for removing redundant information from a GIF or MNG animation.
The sequence of images is terminated by the appearance of any option. If the -deconstruct option appears after all of the input images, all images are deconstructed.
The following definitions may be created:
cineon:colorspace={rgb|cineonlog}
Use the cineon:colorspace option when reading a Cineon file to specify the
colorspace the Cineon file uses. This overrides the colorspace type
implied by the DPX header (if any).
dpx:bits-per-sample=<value>
If the dpx:bits-per-sample key is defined, GraphicsMagick will write DPX
images with the specified bits per sample, overriding any existing depth
value. If this option is not specified, then the value is based on the
existing image depth value from the original image file. The DPX
standard supports bits per sample values of 1, 8, 10, 12, and 16. Many
DPX readers demand a sample size of 10 bits with type A padding (see
below).
dpx:colorspace={rgb|cineonlog}
Use the dpx:colorspace option when reading a DPX file to specify the
colorspace the DPX file uses. This overrides the colorspace type implied
by the DPX header (if any).
dpx:packing-method={packed|a|b|lsbpad|msbpad}
DPX samples are output within 32-bit words. They may be tightly packed
end-to-end within the words ("packed"), padded with null bits
to the right of the sample ("a" or "lsbpad"), or
padded with null bits to the left of the sample ("b" or
"msbpad"). This option only has an effect for sample sizes of
10 or 12 bits. If samples are not packed, the DPX standard recommends
type A padding. Many DPX readers demand a sample size of 10 bits with
type A padding.
dpx:pixel-endian={lsb|msb}
Allows the user to specify the endian order of the pixels when reading or
writing the DPX files. Sometimes this is useful if the file is (or must
be) written incorrectly so that the file header and the pixels use
different endianness.
dpx:swap-samples={true|false}
dpx:swap-samples-read={true|false}
dpx:swap-samples-write={true|false}
GraphicsMagick strives to adhere to the DPX standard but certain aspects
of the standard can be quite confusing. As a result, some 10-bit DPX
files have Red and Blue interchanged, or Cb and Cr interchanged due to
an different interpretation of the standard, or getting the wires
crossed. The swap-samples option may be supplied when reading or writing
in order to read or write using the necessary sample order. Use
swap-samples-read when swapping should only occur in the reader, or
swap-samples-write when swapping should only occur in the writer.
gradient:direction={South|North|West|East|NorthWest|NorthEast|SouthWest|SouthEast}
By default, the gradient coder produces a gradient from top to bottom
("South"). Since GraphicsMagick 1.3.35, the gradient direction
may be specified to produce gradient vectors according to a gravity-like
specification. The arguments are South (Top to Bottom),
North (Bottom to Top), West (Right to Left), East
(Left to Right), NorthWest (Bottom-Right to Top-Left),
NorthEast (Bottom-Left to Top-Right), SouthWest (Top-Right
Bottom-Left), and SouthEast (Top-Left to Bottom-Right).
jp2:rate=<value>
Specify the compression factor to use while writing JPEG-2000 files. The
compression factor is the reciprocal of the compression ratio. The valid
range is 0.0 to 1.0, with 1.0 indicating lossless compression. If
defined, this value overrides the -quality setting. The default quality
setting of 75 results in a rate value of 0.06641.
jpeg:arithmetic={true|false}
Enables or disables arithmetic encoding if the JPEG library supports it
(default disabled). When this is enabled, optimize-coding is necessarily
disabled.
jpeg:block-smoothing={true|false}
Enables or disables block smoothing when reading a JPEG file (default
enabled).
jpeg:dct-method=<value>
Selects the IJG JPEG library DCT implementation to use. The encoding
implementations vary in speed and encoding error. The available choices
for value are islow, ifast, float,
default and fastest. Note that fastest might not
necessarily be fastest on your CPU, depending on the choices made when
the JPEG library was built and how your CPU behaves.
jpeg:fancy-upsampling={true|false}
Enables or disables fancy upsampling when reading a JPEG file (default
enabled).
jpeg:max-scan-number=<value>
Specifies an integer value for the maximum number of progressive scans
allowed in a JPEG file. The default maximum is 100 scans. This limit is
imposed due to a weakness in the JPEG standard which allows small JPEG
files to take many minutes or hours to be read.
jpeg:max-warnings=<value>
Specifies an integer value for how many warnings are allowed for any given
error type before being promoted to a hard error. JPEG files producing
excessive warnings indicate a problem with the file.
jpeg:optimize-coding={true|false}
Selects if huffman encoding should be used. Huffman encoding is enabled by
default, but may be disabled for very large images since it encoding
requires that the entire image be buffered in memory. Huffman encoding
produces smaller JPEG files at the expense of added compression time and
memory consumption.
jpeg:preserve-settings
If the jpeg:preserve-settings flag is defined, the JPEG encoder will use
the same "quality" and "sampling-factor" settings
that were found in the input file, if the input was in JPEG format.
These settings are also preserved if the input is a JPEG file and the
output is a JNG file. If the colorspace of the output file differs from
that of the input file, the quality setting is preserved but the
sampling-factors are not.
pcl:fit-to-page
If the pcl:fit-to-page flag is defined, then the printer is requested to
scale the image to fit the page size (width and/or height).
png:chunk-malloc-max=<value>
png:chunk-malloc-max specifies the maximum chunk size that libpng will be
allowed to read. Libpng's default is normally 8,000,000 bytes. Very
rarely, a valid PNG file may be encountered where the error is reported
"chunk data is too large". In this case, the limit may be
increased using this option. Take care when increasing this limit since
an excessively large limit could allow untrusted files to use excessive
memory.
mng:maximum-loops=<value>
mng:maximum-loops specifies the maximum number of loops allowed to be
specified by a MNG LOOP chunk. Without an imposed limit, a MNG file
could request up to 2147483647 loops, which could run for a very long
time. The current default limit is 512 loops.
pdf:use-cropbox={true|false}
If the pdf:use-cropbox flag is set to true, then Ghostscript is
requested to apply the PDF crop box.
pdf:stop-on-error={true|false}
If the pdf:stop-on-error flag is set to true, then Ghostscript is
requested to stop processing the PDF when the first error is
encountered. Otherwise it will attempt to process all requested
pages.
ps:imagemask
If the ps:imagemask flag is defined, the PS3 and EPS3 coders will create
Postscript files that render bilevel images with the Postscript
imagemask operator instead of the image operator.
ptif:minimum-geometry=<geometry>
If the ptif:minimum-geometry key is defined, GraphicsMagick will use it to
determine the minimum frame size to output when writing a pyramid TIFF
file (a TIFF file containing a succession of reduced versions of the
first frame). The default minimum geometry is 32x32.
tiff:alpha={unspecified|associated|unassociated}
Specify the TIFF alpha channel type when reading or writing TIFF files,
overriding the normal value. The default alpha channel type for new
files is unspecified alpha. Existing alpha settings are preserved when
converting from one TIFF file to another. When a TIFF file uses
associated alpha, the image pixels are pre-multiplied (i.e. altered)
with the alpha channel. Files with "associated" alpha appear
as if they were alpha composited on a black background when the matte
channel is disabled. If the unassociated alpha type is selected, then
the alpha channel is saved without altering the pixels. Photoshop
recognizes associated alpha as transparency information, if the file is
saved with unassociated alpha, the alpha information is loaded as an
independent channel. Note that for many years, ImageMagick and
GraphicsMagick marked TIFF files as using associated alpha, without
properly pre-multiplying the pixels.
tiff:fill-order={msb2lsb|lsb2msb}
If the tiff:fill-order key is defined, GraphicsMagick will use it to
determine the bit fill order used while writing TIFF files. The normal
default is "msb2lsb", which matches the native bit order of
all modern CPUs. The only exception to this is when Group3 or Group4 FAX
compression is requested since FAX machines send data in bit-reversed
order and therefore RFC 2301 recommends using reverse order.
tiff:group-three-options=<value>
If the tiff:group-three-options key is defined, GraphicsMagick will use it
to set the group3 options tag when writing group3-compressed TIFF.
Please see the TIFF specification for the usage of this tag. The default
value is 4.
tiff:ignore-tags=<tags>
If the tiff:ignore-tags key is defined, then it is used as a list of
comma-delimited integer TIFF tag values to ignore while reading the TIFF
file. This is useful in order to be able to read files which which
otherwise fail to read due to problems with TIFF tags. Note that some
TIFF tags are required in order to be able to read the image data at
all.
tiff:report-warnings={false|true}
If the tiff:report-warnings key is defined and set to true, then
TIFF warnings are reported as a warning exception rather than as a coder
log message. Such warnings are reported after the image has been read or
written. Most TIFF warnings are benign but sometimes they may help
deduce problems with the TIFF file, or help detect that the TIFF file
requires a special application to read successfully due to the use of
proprietary or specialized extensions.
tiff:sample-format={unsigned|ieeefp}
If the tiff:sample-format key is defined, GraphicsMagick will use it to
determine the sample format used while writing TIFF files. The default
is "unsigned". Specify "ieeefp" in order to write
floating-point TIFF files with float (32-bit) or double (64-bit) values.
Use the tiff:bits-per-sample define to determine the type of
floating-point value to use.
tiff:max-sample-value=<value>
If the tiff:max-sample-value key is defined, GraphicsMagick will use the
assigned value as the maximum floating point value while reading or
writing IEEE floating point TIFFs. Otherwise the maximum value is 1.0 or
the value obtained from the file's SMaxSampleValue tag (if present). The
floating point data is currently not scanned in advance to determine a
best maximum sample value so if the range is not 1.0, or the
SMaxSampleValue tag is not present, it may be necessary to
(intelligently) use this parameter to properly read a file.
tiff:min-sample-value=<value>
If the tiff:min-sample-value key is defined, GraphicsMagick will use the
assigned value as the minimum floating point value while reading or
writing IEEE floating point TIFFs. Otherwise the minimum value is 0.0 or
the value obtained from the file's SMinSampleValue tag (if present).
tiff:bits-per-sample=<value>
If the tiff:bits-per-sample key is defined, GraphicsMagick will write
images with the specified bits per sample, overriding any existing depth
value. Value may be any in the range of 1 to 32, or 64 when the default
´unsigned' format is written, or 16/32/24/64 if IEEEFP format is
written. Please note that the baseline TIFF 6.0 specification only
requires readers to handle certain powers of two, and the values to be
handled depend on the nature of the image (e.g. colormapped, grayscale,
RGB, CMYK).
tiff:samples-per-pixel=<value>
If the tiff:samples-per-pixel key is defined to a value, the TIFF coder
will write TIFF images with the defined samples per pixel, overriding
any value stored in the image. This option should not normally be
used.
tiff:rows-per-strip=<value>
Allows the user to specify the number of rows per TIFF strip. Rounded up
to a multiple of 16 when using JPEG compression. Ignored when using
tiles.
tiff:strip-per-page=true
Requests that the image is written in a single TIFF strip. This is
normally the default when group3 or group4 compression is requested
within reasonable limits. Requesting a single strip for large images may
result in failure due to resource consumption in the writer or
reader.
tiff:tile
Enable writing tiled TIFF (rather than stripped) using the default tile
size. Tiled TIFF organizes the image as an array of smaller images
(tiles) in order to enable random access.
tiff:tile-geometry=<width>x<height>
Specify the tile size to use while writing tiled TIFF. Width and height
should be a multiple of 16. If the value is not a multiple of 16, then
it will be rounded down. Enables tiled TIFF if it has not already been
enabled. GraphicsMagick does not use tiled storage internally so tiles
need to be converted back and forth from the internal scanline-oriented
storage to tile-oriented storage. Testing with typical RGB images shows
that useful square tile size values range from 128x128 to 1024x1024.
Large images which require using a disk-based pixel cache benefit from
large tile sizes while images which fit in memory work well with smaller
tile sizes.
tiff:tile-width=<width>
Specify the tile width to use while writing tiled TIFF. The tile height is
then defaulted to an appropriate size. Width should be a multiple of 16.
If the value is not a multiple of 16, then it will be rounded down.
Enables tiled TIFF if it has not already been enabled.
tiff:tile-height=<height>
Specify the tile height to use while writing tiled TIFF. The tile width is
then defaulted to an appropriate size. Height should be a multiple of
16. If the value is not a multiple of 16, then it will be rounded down.
Enables tiled TIFF if it has not already been enabled.
tiff:webp-lossless={TRUE|FALSE}
Specify a value of TRUE to enable lossless mode while writing
WebP-compressed TIFF files. The WebP webp:lossless option may
also be used. The quality factor set by the -quality option may
be used to influence the level of effort expended while compressing.
tiff:zstd-compress-level=<value>
Specify the compression level to use while writing Zstd-compressed TIFF
files. The valid range is 1 to 22. If this define is not specified, then
the 'quality' value is used such that the default quality setting of 75
is translated to a compress level of 9 such that ´quality' has a
useful range of 10-184 if used for this purpose.
webp:lossless={true|false}
Enable lossless encoding.
webp:method={0-6}
Quality/speed trade-off.
webp:image-hint={default,graph,photo,picture}
Hint for image type.
webp:target-size=<integer>
Target size in bytes.
webp:target-psnr=<float>
Minimal distortion to try to achieve.
webp:segments={1-4}
Maximum number of segments to use.
webp:sns-strength={0-100}
Spatial Noise Shaping.
webp:filter-strength={0-100}
Filter strength.
webp:filter-sharpness={0-7}
Filter sharpness.
webp:filter-type={0,1}
Filtering type. 0 = simple, 1 = strong (only used if filter-strength >
0 or autofilter is enabled).
webp:auto-filter={true|false}
Auto adjust filter's strength.
webp:alpha-compression=<integer>
Algorithm for encoding the alpha plane (0 = none, 1 = compressed with WebP
lossless). Default is 1.
webp:alpha-filtering=<integer>
Predictive filtering method for alpha plane. 0: none, 1: fast, 2: best.
Default is 1.
webp:alpha-quality={0-100}
Between 0 (smallest size) and 100 (lossless). Default is 100.
webp:pass=[1..10]
Number of entropy-analysis passes.
webp:show-compressed={true|false}
Export the compressed picture back. In-loop filtering is not applied.
webp:preprocessing=[0,1,2]
0=none, 1=segment-smooth, 2=pseudo-random dithering
webp:partitions=[0-3]
log2(number of token partitions) in [0..3]. Default is 0 for easier
progressive decoding.
webp:partition-limit={0-100}
Quality degradation allowed to fit the 512k limit on prediction modes
coding (0: no degradation, 100: maximum possible degradation).
webp:emulate-jpeg-size={true|false}
If true, compression parameters will be remapped to better match the
expected output size from JPEG compression. Generally, the output size
will be similar but the degradation will be lower.
webp:thread-level=<integer>
If non-zero, try and use multi-threaded encoding.
webp:low-memory={true|false}
If set, reduce memory usage (but increase CPU use)
webp:use-sharp-yuv={true|false}
If set, if needed, use sharp (and slow) RGB->YUV conversion
For example, to create a postscript file that will render only the black pixels of a bilevel image, use:
gm convert bilevel.tif -define ps:imagemask eps3:stencil.ps
This option is useful for regulating the animation of image sequences Delay/100 seconds must expire before the display of the next image. The default is no delay between each showing of the image sequence. The maximum delay is 65535.
You can specify a delay range (e.g. -delay 10-500) which sets the minimum and maximum delay.
This is the number of bits of color to preserve in the image. Any value between 1 and QuantumDepth (build option) may be specified, although 8 or 16 are the most common values. Use this option to specify the depth of raw images whose depth is unknown such as GRAY, RGB, or CMYK, or to change the depth of any image after it has been read. The depth option is applied to the pixels immediately so it may be used as a form of simple compression by discarding the least significant bits. Reducing the depth in advance may speed up color quantization, and help create smaller file sizes when using a compression algorithm like LZW or ZIP.
With this option, composite image is used as a displacement map. Black, within the displacement map, is a maximum positive displacement. White is a maximum negative displacement and middle gray is neutral. The displacement is scaled to determine the pixel shift. By default, the displacement applies in both the horizontal and vertical directions. However, if you specify mask, composite image is the horizontal X displacement and mask the vertical Y displacement.
This option is used with convert for obtaining image or font from this X server. See X(1).
The Disposal Method indicates the way in which the graphic is to be treated after being displayed.
Here are the valid methods:
Undefined No disposal specified.
None Do not dispose between frames.
Background Overwrite the image area with
the background color.
Previous Overwrite the image area with
what was there prior to rendering
the image.
The opacity of the composite image is multiplied by the given percent, then it is composited over the main image.
The basic strategy of dithering is to trade intensity resolution for spatial resolution by averaging the intensities of several neighboring pixels. Images which suffer from severe contouring when reducing colors can be improved with this option.
The -colors or -monochrome option is required for this option to take effect.
Use +dither to turn off dithering and to render PostScript without text or graphic aliasing. Disabling dithering often (but not always) leads to decreased processing time.
Use this option to annotate an image with one or more graphic primitives. The primitives include shapes, text, transformations, and pixel operations. The shape primitives are
point x,y
line x0,y0 x1,y1
rectangle x0,y0 x1,y1
roundRectangle x0,y0 x1,y1 wc,hc
arc x0,y0 x1,y1 a0,a1
ellipse x0,y0 rx,ry a0,a1
circle x0,y0 x1,y1
polyline x0,y0 ... xn,yn
polygon x0,y0 ... xn,yn
Bezier x0,y0 ... xn,yn
path path specification
image operator x0,y0 w,h filename
The text primitive is
text x0,y0 string
The text gravity primitive is
gravity NorthWest, North, NorthEast, West, Center,
East, SouthWest, South, or SouthEast
The text gravity primitive only affects the placement of text and does not interact with the other primitives. It is equivalent to using the -gravity commandline option, except that it is limited in scope to the -draw option in which it appears.
The transformation primitives are
rotate degrees
translate dx,dy
scale sx,sy
skewX degrees
skewY degrees
The pixel operation primitives are
color x0,y0 method
matte x0,y0 method
The shape primitives are drawn in the color specified in the preceding -stroke option. Except for the line and point primitives, they are filled with the color specified in the preceding -fill option. For unfilled shapes, use -fill none.
Point requires a single coordinate.
Line requires a start and end coordinate.
Rectangle expects an upper left and lower right coordinate.
RoundRectangle has the upper left and lower right coordinates and the width and height of the corners.
Circle has a center coordinate and a coordinate for the outer edge.
Use Arc to inscribe an elliptical arc within a rectangle. Arcs require a start and end point as well as the degree of rotation (e.g. 130,30 200,100 45,90).
Use Ellipse to draw a partial ellipse centered at the given point with the x-axis and y-axis radius and start and end of arc in degrees (e.g. 100,100 100,150 0,360).
Finally, polyline and polygon require three or more coordinates to define its boundaries. Coordinates are integers separated by an optional comma. For example, to define a circle centered at 100,100 that extends to 150,150 use:
-draw 'circle 100,100 150,150'
Paths (See Paths) represent an outline of an object which is defined in terms of moveto (set a new current point), lineto (draw a straight line), curveto (draw a curve using a cubic Bezier), arc (elliptical or circular arc) and closepath (close the current shape by drawing a line to the last moveto) elements. Compound paths (i.e., a path with subpaths, each consisting of a single moveto followed by one or more line or curve operations) are possible to allow effects such as "donut holes" in objects.
Use image to composite an image with another image. Follow the image keyword with the composite operator, image location, image size, and filename:
-draw 'image Over 100,100 225,225 image.jpg'
You can use 0,0 for the image size, which means to use the actual dimensions found in the image header. Otherwise, it will be scaled to the given dimensions. See -compose for a description of the composite operators.
Use text to annotate an image with text. Follow the text coordinates with a string. If the string has embedded spaces, enclose it in single or double quotes. Optionally you can include the image filename, type, width, height, or other image attribute by embedding special format character. See -comment for details.
For example,
-draw 'text 100,100 "%m:%f %wx%h"' annotates the image with MIFF:bird.miff 512x480 for an image titled bird.miff and whose width is 512 and height is 480. If the first character of string is @, the text is read from a file titled by the remaining characters in the string. Please note that if the string comes from an untrusted source that it should be sanitized before use (a security risk). Rotate rotates subsequent shape primitives and text primitives about the origin of the main image. If the -region option precedes the -draw option, the origin for transformations is the upper left corner of the region. Translate translates them. Scale scales them. SkewX and SkewY skew them with respect to the origin of the main image or the region. The transformations modify the current affine matrix, which is initialized from the initial affine matrix defined by the -affine option. Transformations are cumulative within the -draw option. The initial affine matrix is not affected; that matrix is only changed by the appearance of another -affine option. If another -draw option appears, the current affine matrix is reinitialized from the initial affine matrix. Use color to change the color of a pixel to the fill color (see -fill). Follow the pixel coordinate with a method:
point
replace
floodfill
filltoborder
reset
Consider the target pixel as that specified by your coordinate. The point method recolors the target pixel. The replace method recolors any pixel that matches the color of the target pixel. Floodfill recolors any pixel that matches the color of the target pixel and is a neighbor, whereas filltoborder recolors any neighbor pixel that is not the border color. Finally, reset recolors all pixels.
Use matte to the change the pixel matte value to transparent. Follow the pixel coordinate with a method (see the color primitive for a description of methods). The point method changes the matte value of the target pixel. The replace method changes the matte value of any pixel that matches the color of the target pixel. Floodfill changes the matte value of any pixel that matches the color of the target pixel and is a neighbor, whereas filltoborder changes the matte value of any neighbor pixel that is not the border color (-bordercolor). Finally reset changes the matte value of all pixels.
You can set the primitive color, font, and font bounding box color with -fill, -font, and -box respectively. Options are processed in command line order so be sure to use these options before the -draw option.
Choose from AdobeCustom, AdobeExpert, AdobeStandard, AppleRoman, BIG5, GB2312, Latin 2, None, SJIScode, Symbol, Unicode, Wansung.
MSB indicates big-endian (e.g. SPARC, Motorola 68K) while LSB indicates little-endian (e.g. Intel 'x86, VAX) byte ordering. Native indicates to use the normal ordering for the current CPU. This option currently only influences the CMYK, DPX, GRAY, RGB, and TIFF, formats.
Use +endian to revert to unspecified endianness.
This option composites the image on a new background color (-background) canvas image of size <width>x<height>. The existing image content is composited at the position specified by geometry x and y offset and/or desired gravity (-gravity) using the current image compose (-compose) method. Image content which falls outside the bounds of the new image dimensions is discarded.
For example, this command creates a thumbnail of an image, and centers it on a red color backdrop image, offsetting the canvas ten pixels to the left and five pixels up, with respect to the thumbnail:
gm convert infile.jpg -thumbnail 120x80 -background red -gravity center \
-extent 140x100-10-5 outfile.jpg
This command reduces or expands a JPEG image to fit on an 800x600 display:
gm convert -size 800x600 input.jpg \
-resize 800x600 -background black \
-compose Copy -gravity center \
-extent 800x600 \
-quality 92 output.jpg
If the aspect ratio of the input image isn't exactly 4:3, then the image is centered on an 800x600 black canvas.
If -file is specified, then an annotated difference image is generated and written to the specified file. Pixels which differ between the reference and compare images are modified from those in the compare image so that the changed pixels become more obvious. Some images may require use of an alternative highlight style (see -highlight-style) or highlight color (see -highlight-color) before the changes are obvious.
Colors are represented in GraphicsMagick in the same form used by SVG. Use "gm convert -list color" to list named colors:
name (named color)
#RGB (hex numbers, 4 bits each)
#RRGGBB (8 bits each)
#RRRGGGBBB (12 bits each)
#RRRRGGGGBBBB (16 bits each)
#RGBA (4 bits each)
#RRGGBBAA (8 bits each)
#RRRGGGBBBAAA (12 bits each)
#RRRRGGGGBBBBAAAA (16 bits each)
rgb(r,g,b) (r,g,b are decimal numbers)
rgba(r,g,b,a) (r,g,b,a are decimal numbers)
Enclose the color specification in quotation marks to prevent the "#" or the parentheses from being interpreted by your shell.
For example,
gm convert -fill blue ...
gm convert -fill "#ddddff" ...
gm convert -fill "rgb(65000,65000,65535)" ...
The shorter forms are scaled up, if necessary by replication. For example, #3af, #33aaff, and #3333aaaaffff are all equivalent.
See -draw for further details.
Use this option to affect the resizing operation of an image (see -geometry). Choose from these filters (ordered by approximate increasing CPU time):
Point
Box
Triangle
Hermite
Hanning
Hamming
Blackman
Gaussian
Quadratic
Cubic
Catrom
Mitchell
Lanczos
Bessel
Sinc
The default filter is automatically selected to provide the best quality while consuming a reasonable amount of time. The Mitchell filter is used if the image supports a palette, supports a matte channel, or is being enlarged, otherwise the Lanczos filter is used.
In some file formats (e.g. Photoshop's PSD) complex images may be represented by "layers" (independent images) which must be composited in order to obtain the final rendition. The -flatten option accomplishes this composition. The sequence of images is replaced by a single image created by compositing each image in turn, while respecting composition operators and page offsets. While -flatten is immediately useful for eliminating layers, it is also useful as a general-purpose composition tool.
The sequence of images is terminated by the appearance of any option. If the -flatten option appears after all of the input images, all images are flattened. Also see -mosaic which is similar to -flatten except that it adds a suitably-sized canvas base image.
For example, this composites an image on top of a 640x400 transparent black canvas image:
gm convert -size 640x300 xc:transparent \
-compose over -page +0-100 \
frame.png -flatten output.png
and this flattens a Photoshop PSD file:
gm convert input.psd -flatten output.png
reflect the scanlines in the vertical direction.
reflect the scanlines in the horizontal direction.
You can tag a font to specify whether it is a PostScript, TrueType, or X11 font. For example, Arial.ttf is a TrueType font, ps:helvetica is PostScript, and x:fixed is X11.
The color is specified using the format described under the -fill option.
When used with the mogrify utility, this option will convert any image to the image format you specify. See GraphicsMagick(1) for a list of image format types supported by GraphicsMagick, or see the output of 'gm -list format'.
By default the file is written to its original name. However, if the filename extension matches a supported format, the extension is replaced with the image format type specified with -format. For example, if you specify tiff as the format type and the input image filename is image.gif, the output image filename becomes image.tiff.
When used with the identify utility, or the convert utility with output written to the 'info:-' file specification, use this option to print information about the image in a format of your choosing. You can include the image filename, type, width, height, Exif data, or other image attributes by embedding special format characters:
%b file size
%c comment
%d directory
%e filename extension
%f filename
%g page dimensions and offsets
%h height
%i input filename
%k number of unique colors
%l label
%m magick
%n number of scenes
%o output filename
%p page number
%q image bit depth
%r image type description
%s scene number
%t top of filename
%u unique temporary filename
%w width
%x horizontal resolution
%y vertical resolution
%A transparency supported
%C compression type
%D GIF disposal method
%G Original width and height
%H page height
%M original filename specification
%O page offset (x,y)
%P page dimensions (width,height)
%Q compression quality
%T time delay (in centi-seconds)
%U resolution units
%W page width
%X page horizontal offset (x)
%Y page vertical offset (y)
%@ trim bounding box
%# signature
\n newline
\r carriage return
%% %
For example,
-format "%m:%f %wx%h"
displays MIFF:bird.miff 512x480 for an image titled bird.miff and whose width is 512 and height is 480.
If the first character of string is @, the format is read from a file titled by the remaining characters in the string. Please note that if the string comes from an untrusted source that it should be sanitized before use since this may be used to incorporate any readable file on the system (a security risk).
The values of image type (%r) which may be returned include:
Bilevel
Grayscale
GrayscaleMatte
Palette
PaletteMatte
TrueColor
TrueColorMatte
ColorSeparation
ColorSeparationMatte
Optimize
You can also use the following special formatting syntax to print Exif information contained in the file:
%[EXIF:<tag>]
Where "<tag>" may be one of the following:
* (print all Exif tags, in keyword=data format)
! (print all Exif tags, in tag_number format)
#hhhh (print data for Exif tag #hhhh)
ImageWidth
ImageLength
BitsPerSample
Compression
PhotometricInterpretation
FillOrder
DocumentName
ImageDescription
Make
Model
StripOffsets
Orientation
SamplesPerPixel
RowsPerStrip
StripByteCounts
XResolution
YResolution
PlanarConfiguration
ResolutionUnit
TransferFunction
Software
DateTime
Artist
WhitePoint
PrimaryChromaticities
TransferRange
JPEGProc
JPEGInterchangeFormat
JPEGInterchangeFormatLength
YCbCrCoefficients
YCbCrSubSampling
YCbCrPositioning
ReferenceBlackWhite
CFARepeatPatternDim
CFAPattern
BatteryLevel
Copyright
ExposureTime
FNumber
IPTC/NAA
ExifOffset
InterColorProfile
ExposureProgram
SpectralSensitivity
GPSInfo
ISOSpeedRatings
OECF
ExifVersion
DateTimeOriginal
DateTimeDigitized
ComponentsConfiguration
CompressedBitsPerPixel
ShutterSpeedValue
ApertureValue
BrightnessValue
ExposureBiasValue
MaxApertureValue
SubjectDistance
MeteringMode
LightSource
Flash
FocalLength
MakerNote
UserComment
SubSecTime
SubSecTimeOriginal
SubSecTimeDigitized
FlashPixVersion
ColorSpace
ExifImageWidth
ExifImageLength
InteroperabilityOffset
FlashEnergy
SpatialFrequencyResponse
FocalPlaneXResolution
FocalPlaneYResolution
FocalPlaneResolutionUnit
SubjectLocation
ExposureIndex
SensingMethod
FileSource
SceneType
JPEG specific information (from reading a JPEG file) may be obtained like this:
%[JPEG-<tag>]
Where "<tag>" may be one of the following:
* (all JPEG-related tags, in
keyword=data format)
Quality IJG JPEG "quality" estimate
Colorspace JPEG colorspace numeric ID
Colorspace-Name JPEG colorspace name
Sampling-factors JPEG sampling factors
Please note that JPEG has no notion of "quality" and that the quality metric used by, and estimated by the software is based on the quality metric established by IJG JPEG 6b. Other encoders (e.g. that used by Adobe Photoshop) use different encoding metrics.
Surround the format specification with quotation marks to prevent your shell from misinterpreting any spaces and square brackets.
See -geometry for details about the geometry specification. The -frame option is not affected by the -gravity option.
The color of the border is specified with the -mattecolor command line option.
A number of algorithms search for a target color. By default the color must be exact. Use this option to match colors that are close (in Euclidean distance) to the target color in RGB 3D space. For example, if you want to automatically trim the edges of an image with -trim but the image was scanned and the target background color may differ by a small amount. This option can account for these differences.
The distance can be in absolute intensity units or, by appending "%", as a percentage of the maximum possible intensity (255, 65535, or 4294967295).
The same color image displayed on two different workstations may look different due to differences in the display monitor. Use gamma correction to adjust for this color difference. Reasonable values extend from 0.8 to 2.3. Gamma less than 1.0 darkens the image and gamma greater than 1.0 lightens it. Large adjustments to image gamma may result in the loss of some image information if the pixel quantum size is only eight bits (quantum range 0 to 255).
You can apply separate gamma values to the red, green, and blue channels of the image with a gamma value list delimited with slashes (e.g., 1.7/2.3/1.2).
Use +gamma value to set the image gamma level without actually adjusting the image pixels. This option is useful if the image is of a known gamma but not set as an image attribute (e.g. PNG images).
Use the given radius and standard deviation (sigma).
The -geometry option is used for a number of different purposes, depending on the utility it is used with.
For the X11 commands ('animate', 'display', and 'import'), it
specifies the preferred size and location of the Image window. By
default, the window size is the image size and the location is chosen by
you (or your window manager) when it is mapped.
For the 'import', 'convert', 'mogrify' utility commands it may be used to
specify the desired size when resizing an image. In this case, symbols
representing resize options may be appended to the geometry string to
influence how the resize request is treated.
See later notes corresponding to usage by particular commands. The following notes apply to when -geometry is used to express a resize request, taking into account the current properties of the image.
By default, the width and height are maximum values. That is, the image is expanded or contracted to fit the width and height value while maintaining the aspect ratio of the image.
Append a ^ to the geometry so that the image aspect ratio is maintained when the image is resized, but the resulting width or height are treated as minimum values rather than maximum values.
Append a ! (exclamation point) to the geometry to force the image size to exactly the size you specify. For example, if you specify 640x480! the image width is set to 640 pixels and height to 480.
If only the width is specified, without the trailing 'x', then height is set to width (e.g., -geometry 100 is the same as -geometry 100x100). If only the width is specified but with the trailing 'x', then width assumes the value and the height is chosen to maintain the aspect ratio of the image. Similarly, if only the height is specified prefixed by 'x' (e.g., -geometry x256), the width is chosen to maintain the aspect ratio.
To specify a percentage width or height instead, append %. The image size is multiplied by the width and height percentages to obtain the final image dimensions. To increase the size of an image, use a value greater than 100 (e.g. 125%). To decrease an image's size, use a percentage less than 100.
Use @ to specify the maximum area in pixels of an image.
Use > to change the dimensions of the image only if its width or height exceeds the geometry specification. < resizes the image only if both of its dimensions are less than the geometry specification. For example, if you specify '640x480>' and the image size is 256x256, the image size does not change. However, if the image is 512x512 or 1024x1024, it is resized to 480x480. Enclose the geometry specification in quotation marks to prevent the < or > from being interpreted by your shell as a file redirection.
When used with animate and display, offsets are handled in the same manner as in X(1) and the -gravity option is not used. If the x is negative, the offset is measured leftward from the right edge of the screen to the right edge of the image being displayed. Similarly, negative y is measured between the bottom edges. The offsets are not affected by "%"; they are always measured in pixels.
When used as a composite option, -geometry gives the dimensions of the image and its location with respect to the composite image. If the -gravity option is present with NorthEast, East, or SouthEast gravity, the x represents the distance from the right edge of the image to the right edge of the composite image. Similarly, if the -gravity option is present with SouthWest, South, or SouthEast gravity, y is measured between the bottom edges. Accordingly, a positive offset will never point in the direction outside of the image. The offsets are not affected by "%"; they are always measured in pixels. To specify the dimensions of the composite image, use the -resize option.
When used as a convert, import or mogrify option, -geometry is synonymous with -resize and specifies the size of the output image. The offsets, if present, are ignored.
When used as a montage option, -geometry specifies the image size and border size for each tile; default is 256x256+0+0. Negative offsets (border dimensions) are meaningless. The -gravity option affects the placement of the image within the tile; the default gravity for this purpose is Center. If the "%" sign appears in the geometry specification, the tile size is the specified percentage of the original dimensions of the first tile. To specify the dimensions of the montage, use the -resize option.
Choices are: NorthWest, North, NorthEast, West, Center, East, SouthWest, South, SouthEast.
The direction you choose specifies where to position the text when annotating the image. For example Center gravity forces the text to be centered within the image. By default, the image gravity is NorthWest. See -draw for more details about graphic primitives. Only the text primitive is affected by the -gravity option.
The -gravity option is also used in concert with the -geometry option and other options that take <geometry> as a parameter, such as the -crop option. See -geometry for details of how the -gravity option interacts with the <x> and <y> parameters of a geometry specification.
When used as an option to composite, -gravity gives the direction that the image gravitates within the composite.
When used as an option to montage, -gravity gives the direction that an image gravitates within a tile. The default gravity is Center for this purpose.
A Hald CLUT ("Color Look-Up Table") is a special square color image which contains a look-up table for red, green, and blue. The size of the Hald CLUT image is determined by its order. The width (and height) of a Hald CLUT is the cube of the order. For example, a Hald CLUT of order 8 is 512x512 pixels (262,144 colors) and of order 16 is 4096x4096 (16,777,216 colors). A special CLUT is the identity CLUT which which causes no change to the input image. In order to use the Hald CLUT, one takes an identity CLUT and adjusts its colors in some way. The modified CLUT can then be used to transform any number of images in an identical way.
GraphicsMagick contains a built-in identity CLUT generator via the IDENTITY coder. For example reading from the file name IDENTITY:8 returns an identity CLUT of order 8. Typical Hald CLUT identity images have an order of between 8 and 16. The default order for the IDENTITY CLUT generator is 8. Interpolation is used so it is not usually necessary for CLUT images to be very large. The PNG file format is ideal for storing Hald CLUT images because it compresses them very well.
Specifies the color to use when annotating difference pixels.
Specifies the pixel difference annotation style used to draw attention to changed pixels. May be one of Assign, Threshold, Tint, or XOR; where Assign replaces the pixel with the highlight color (see -highlight-color), Threshold replaces the pixel with black or white based on the difference in intensity, Tint alpha tints the pixel with the highlight color, and XOR does an XOR between the pixel and the highlight color.
Offsets, if present in the geometry specification, are handled in the same manner as the -geometry option, using X11 style to handle negative offsets.
Use this option to affect the the color management operation of an image (see -profile). Choose from these intents: Absolute, Perceptual, Relative, Saturation.
The default intent is undefined.
Choices are: None, Line, Plane, or Partition. The default is None.
This option is used to specify the type of interlacing scheme for raw image formats such as RGB or YUV. None means do not interlace (RGBRGBRGBRGBRGBRGB...),
Line uses scanline interlacing (RRR...GGG...BBB...RRR...GGG...BBB...), and Plane uses plane interlacing (RRRRRR...GGGGGG...BBBBBB...).
Partition is like plane except the different planes are saved to individual files (e.g. image.R, image.G, and image.B).
Use Line to create an interlaced PNG or GIF or progressive JPEG image.
Use this option to assign a specific label to the image, when writing to an image format that supports labels, such as TIFF, PNG, MIFF, or PostScript. You can include the the image filename, type, width, height, or other image attribute by embedding special format character. A label is not drawn on the image, but is embedded in the image datastream via a "Label" tag or similar mechanism. If you want the label to be visible on the image itself, use the -draw option. See -comment for details.
For example,
-label "%m:%f %wx%h"
produces an image label of MIFF:bird.miff 512x480 for an image titled bird.miff and whose width is 512 and height is 480.
If the first character of string is @, the image label is read from a file titled by the remaining characters in the string. Please note that if the string comes from an untrusted source that it should be sanitized before use since otherwise the content of an arbitrary readable file might be incorporated into the image label (a security risk).
If the -label option appears multiple times, only the last label is stored.
In PNG images, the label is stored in a tEXt or zTXt chunk with the keyword "label".
When converting to PostScript, use this option to specify a header string to print above the image. Specify the label font with -font.
When creating a montage, by default the label associated with an image is displayed with the corresponding tile in the montage. Use the +label option to suppress this behavior.
Perform local adaptive thresholding using the specified width, height, and offset. The offset is a distance in sample space from the mean, as an absolute integer ranging from 0 to the maximum sample value or as a percentage. If the percent option is supplied, then the offset is computed as a percentage of the quantum range. It is strongly recommended to use the percent option so that results are not sensitive to pixel quantum depth.
For example,
-colorspace gray -lat "10x10-5%"
will help clarify a scanned grayscale or color document, producing a bi-level equivalent.
Give one, two or three values delimited with commas: black-point, gamma, white-point (e.g. 10,1.0,250 or 2%,0.5,98%). The black and white points range from 0 to MaxRGB or from 0 to 100%; if the white point is omitted it is set to MaxRGB-black_point. If a "%" sign is present anywhere in the string, the black and white points are percentages of MaxRGB. Gamma is an exponent that ranges from 0.1 to 10.; if it is omitted, the default of 1.0 (no gamma correction) is assumed. This interface works similar to Photoshop's "Image->Adjustments->Levels..." "Input Levels" interface.
By default, resource limits are estimated based on the available resources and capabilities of the system. The resource limits are Disk, maximum total disk space consumed; File, maximum number of file descriptors allowed to be open at once; Map, maximum total number of file bytes which may be memory mapped; Memory, maximum total number of bytes of heap memory used for image storage; Pixels, maximum absolute image size (per image); Width, maximum image pixels width; Height, maximum image pixels height; Read, maximum number of uncompressed bytes to read; and Threads, the maximum number of worker threads to use per OpenMP thread team.
The Disk and Map resource limits are used to decide if (for a given image) the decoded image ("pixel cache") should be stored in heap memory (RAM), in a memory-mapped disk file, or in a disk file accessed via read/write I/O.
The number of total pixels in one image (Pixels), and/or the width/height (Width/Height), may be limited in order to force the reading, or creation of images larger than the limit (in pixels) to intentionally fail. The disk limit (Disk) establishes an overall limit since using the disk is the means of last resort. When the disk limit has been reached, no more images may be read.
The amount of uncompressed data read when reading one image may be limited by the Read limit. Reading the image fails when the limit is hit. This option is useful if the data is read from a stream (pipe) or from a compressed file such as a gzipped file. Some files are very compressable and so a small compressed file can decompress to a huge amount of data. This option also defends against files which produce seemingly endless loops while decoding by seeking backwards in the file.
The value argument is an absolute value, but may have standard binary suffix characters applied ('K', 'M', 'G', 'T', 'P', 'E') to apply a scaling to the value (based on a multiplier of 1024). Any additional characters are ignored. For example, '-limit Pixels 10MP' limits the maximum image size to 10 megapixels and '-limit memory 32MB -limit map 64MB' limits memory and memory mapped files to 32 megabytes and 64 megabytes respectively.
Resource limits may also be set using environment variables. The environment variables MAGICK_LIMIT_DISK, MAGICK_LIMIT_FILES, MAGICK_LIMIT_MAP, MAGICK_LIMIT_MEMORY, MAGICK_LIMIT_PIXELS, MAGICK_LIMIT_WIDTH, MAGICK_LIMIT_HEIGHT. MAGICK_LIMIT_READ, and OMP_NUM_THREADS may be used to set the limits for disk space, open files, memory mapped size, heap memory, per-image pixels, image width, image height, and threads respectively.
Use the option -list resource list the current limits.
Choices are: Color, Delegate, Format, Magic, Module, Resource, or Type. The Module option is only available if GraphicsMagick was built to support loadable modules.
This option lists information about the GraphicsMagick configuration.
This option specifies the format for the log printed when the -debug option is active.
You can display the following components by embedding special format characters:
%d domain
%e event
%f function
%l line
%m module
%p process ID
%r real CPU time
%t wall clock time
%u user CPU time
%% percent sign
\n newline
\r carriage return
For example:
gm convert -debug coders -log "%u %m:%l %e" in.gif out.png
The default behavior is to print all of the components.
A value other than zero forces the animation to repeat itself up to iterations times.
The image size is doubled using linear interpolation.
The displayed image is magnified by factor.
[convert or mogrify]
By default, color reduction chooses an optimal set of colors that best represent the original image. Alternatively, you can choose a particular set of colors from an image file with this option.
Use +map to reduce all images in the image sequence that follows to a single optimal set of colors that best represent all the images. The sequence of images is terminated by the appearance of any option. If the +map option appears after all of the input images, all images are mapped.
[animate or display]
Choose from these Standard Colormap types:
best
default
gray
red
green
blue
The X server must support the Standard Colormap you choose, otherwise an error occurs. Use list as the type and display searches the list of colormap types in top-to-bottom order until one is located. See xstdcmap(1) for one way of creating Standard Colormaps.
The image read from the file is used as a clipping mask. It must have the same dimensions as the image being masked.
If the mask image contains an opacity channel, the opacity of each pixel is used to define the mask. Otherwise, the intensity (gray level) of each pixel is used. Unmasked (black) pixels are modified while masked pixels (not black) are protected from alteration.
Use +mask to remove the clipping mask.
It is not necessary to use -clip to activate the mask; -clip is implied by -mask.
If the image does not have a matte channel, create an opaque one.
Use +matte to ignore the matte channel (treats it as opaque) and to avoid writing a matte channel in the output file.
For the compare command, -matte will add an opaque matte channel to images if they do not already have a matte channel, and matte will be enabled for both images. Likewise, if +matte is used, the matte channel is disabled for both images. This makes it easier to compare images regardless of if they already have a matte channel.
The color is specified using the format described under the -fill option.
Specifies the maximum amount of total image error (based on comparison using a specified metric) before an error ("image difference exceeds limit") is reported. The error is reported via a non-zero command execution return status.
The image size is halved using linear interpolation.
The available montage modes are frame to place the images in a rectangular grid while adding a decorative frame with dropshadow, unframe to place undecorated images in a rectangular grid, and concatenate to pack the images closely together without any well-defined grid or decoration.
Specify the percent change in brightness, color saturation, and hue separated by commas. Default argument values are 100 percent, resulting in no change. For example, to increase the color brightness by 20% and decrease the color saturation by 10% and leave the hue unchanged, use: -modulate 120,90.
Hue is the percentage of absolute rotation from the current position. For example 50 results in a counter-clockwise rotation of 90 degrees, 150 results in a clockwise rotation of 90 degrees, with 0 and 200 both resulting in a rotation of 180 degrees.
A simple command-line progress indication is shown while the command is running. The process indication shows the operation currently being performed and the percent completed. Commands using X11 may replace the command line progress indication with a graphical one once an image has been displayed.
Both the image pixels and size are linearly interpolated to give the appearance of a meta-morphosis from one image to the next.
The sequence of images is terminated by the appearance of any option. If the -morph option appears after all of the input images, all images are morphed.
The -mosaic option provides a flexible way to composite one or more images onto a solid-color canvas image. It works similar to -flatten except that a base canvas image is automatically created with a suitable size given the image size, page dimensions, and page offsets of images to be composited. The color of the base canvas image may be set via the -background option. The default canvas color is 'white', but 'black' or 'transparent' may be more suitable depending on the composition algorithm requested.
The -compose option may be used to specify the composition algorithm to use when compositing the subsequent image on the base canvas.
The -page option can be used to establish the dimensions of the mosaic and to position the subsequent image within the mosaic. If the -page argument does not specify width and height, then the canvas dimensions are evaluated based on the image sizes and offsets.
The sequence of images is terminated by the appearance of any option. If the -mosaic option appears after all of the input images, all images are included in the mosaic.
The following is an example of composing an image based on red, green, and blue layers extracted from a sequence of images and pasted on the canvas image at specified offsets:
gm convert -background black \
-compose CopyRed -page +0-100 red.png \
-compose CopyGreen -page +0+40 green.png \
-compose CopyBlue -page +0+180 blue.png \
-mosaic output.png
Simulate motion blur by convolving the image with a Gaussian operator of the given radius and standard deviation (sigma). For reasonable results, radius should be larger than sigma. If radius is zero, then a suitable radius is automatically selected based on sigma. The angle specifies the angle that the object is coming from (side which is blurred).
The red, green, and blue intensities of an image are negated. White becomes black, yellow becomes blue, etc. Use +negate to only negate the grayscale pixels of the image.
The principal function of noise peak elimination filter is to smooth the objects within an image without losing edge information and without creating undesired structures. The central idea of the algorithm is to replace a pixel with its next neighbor in value within a pixel window, if this pixel has been found to be noise. A pixel is defined as noise if and only if this pixel is a maximum or minimum within the pixel window.
Use radius to specify the width of the neighborhood.
Use +noise followed by a noise type to add noise to an image. The noise added modulates the existing image pixels. Choose from these noise types:
Uniform
Gaussian
Multiplicative
Impulse
Laplacian
Poisson
Random (uniform distribution)
The -noop option can be used to terminate a group of images and reset all options to their default values, when no other option is desired.
This is a contrast enhancement technique based on the image histogram.
When computing the contrast enhancement values, the histogram edges are truncated so that the majority of the image pixels are considered in the constrast enhancement, and outliers (e.g. random noise or minute details) are ignored. The default is that 0.1 percent of the histogram entries are ignored. The percentage of the histogram to ignore may be specified by using the -set option with the histogram-threshold parameter similar to -set histogram-threshold 0.01 to specify 0.01 percent. Use 0 percent to use the entire histogram, with possibly diminished contrast enhancement.
The color is specified using the format described under the -fill option. The color is replaced if it is identical to the target color, or close enough to the target color in a 3D space as defined by the Euclidean distance specified by -fuzz.
See -fill and -fuzz for more details.
Apply a low-level mathematical, bitwise, or value operator to a selected image channel or all image channels. Operations which result in negative results are reset to zero, and operations which overflow the available range are reset to the maximum possible value.
Select a channel from: Red, Green, Blue, Opacity, Matte, Cyan, Magenta, Yellow, Black, All, or Gray. All only modifies the color channels and does not modify the Opacity channel. Except for the threshold operators, All operates on each channel independently so that operations are on a per-channel basis.
Gray treats the color channels as a grayscale intensity and performs the requested operation on the equivalent pixel intensity so the result is a gray image. Select an operator from Add, And, Assign, Depth, Divide, Gamma, Negate, LShift, Log, Max, Min, Multiply, Or, Pow, RShift, Subtract, Threshold, Threshold-White, Threshold-White-Negate, Threshold-Black, Threshold-Black-Negate, Xor, Noise-Gaussian, Noise-Impulse, Noise-Laplacian, Noise-Multiplicative, Noise-Poisson, Noise-Random, and Noise-Uniform.
Rvalue may be any floating point or integer value. Normally rvalue will be in the range of 0 to MaxRGB, where MaxRGB is the largest quantum value supported by the GraphicsMagick build (255, 65535, or 4294967295) but values outside this range are useful for some arithmetic operations. Arguments to logical or bit-wise operations are rounded to a positive integral value prior to use. If a percent (%) symbol is appended to the argument, then the argument has a range of 0 to 100 percent.
The following is a description of the operators:
Add
Result is rvalue added to channel value.
And
Result is the logical AND of rvalue with channel value.
Assign
Result is rvalue.
Depth
Result is channel value adjusted so that it may be (approximately) stored
in the specified number of bits without additional loss.
Divide
Result is channel value divided by rvalue.
Gamma
Result is channel value gamma adjusted by rvalue.
LShift
Result is channel value bitwise left shifted by rvalue bits.
Log
Result is computed as log(value*rvalue+1)/log(rvalue+1).
Max
Result is assigned to rvalue if rvalue is greater than value.
Min
Result is assigned to rvalue if rvalue is less than value.
Multiply
Result is channel value multiplied by rvalue.
Negate
Result is inverse of channel value (like a film negative). An rvalue must
be supplied but is currently not used. Inverting the image twice results
in the original image.
Or
Result is the logical OR of rvalue with channel value.
Pow
Result is computed as pow(value,rvalue). Similar to Gamma except that
rvalue is not inverted.
RShift
Result is channel value bitwise right shifted by rvalue bits.
Subtract
Result is channel value minus rvalue.
Threshold
Result is maximum (white) if channel value is greater than rvalue, or
minimum (black) if it is less than or equal to rvalue. If all
channels are specified, then thresholding is done based on computed
pixel intensity.
Threshold-white
Result is maximum (white) if channel value is greater than rvalue and is
unchanged if it is less than or equal to rvalue. This can be used to
remove apparent noise from the bright parts of an image. If all
channels are specified, then thresholding is done based on computed
pixel intensity.
Threshold-White-Negate
Result is set to black if channel value is greater than rvalue and is
unchanged if it is less than or equal to rvalue. If all channels
are specified, then thresholding is done based on computed pixel
intensity.
Threshold-black
Result is minimum (black) if channel value is less than than rvalue and is
unchanged if it is greater than or equal to rvalue. This can be used to
remove apparent noise from the dark parts of an image. If all
channels are specified, then thresholding is done based on computed
pixel intensity.
Threshold-Black-Negate
Result is set to white if channel value is less than than rvalue and is
unchanged if it is greater than or equal to rvalue. If all
channels are specified, then thresholding is done based on computed
pixel intensity.
Xor
Result is the logical XOR of rvalue with channel value. An interesting
property of XOR is that performing the same operation twice results in
the original value.
Noise-Gaussian
Result is the current channel value modulated with gaussian noise
according to the intensity specified by rvalue.
Noise-Impulse
Result is the current channel value modulated with impulse noise according
to the intensity specified by rvalue.
Noise-Laplacian
Result is the current channel value modulated with laplacian noise
according to the intensity specified by rvalue.
Noise-Multiplicative
Result is the current channel value modulated with multiplicative gaussian
noise according to the intensity specified by rvalue.
Noise-Poisson
Result is the current channel value modulated with poisson noise according
to the intensity specified by rvalue.
Noise-Random
Result is the current channel value modulated with random (uniform
distribution) noise according to the intensity specified by rvalue. The
initial noise intensity (rvalue=1.0) is the range of one pixel quantum
span.
Noise-Uniform
Result is the channel value with uniform noise applied according to the
intensity specified by rvalue.
As an example, the Assign operator assigns a fixed value to a channel. For example, this command sets the red channel to the mid-range value:
gm convert in.bmp -operator red assign "50%" out.bmp
The following applies 50% thresholding to the image and returns a gray image:
gm convert in.bmp -operator gray threshold "50%" out.bmp
The channel or channels specified in the channeltype argument are reduced to binary, using an ordered dither method. The choices for channeltype are All, Intensity, Red, Green, Blue, Cyan, Magenta, Yellow, Black, and Opacity
When channeltype is "All", the color samples are dithered into a gray level and then that gray level is stored in the three color channels. Separately, the opacity channel is dithered into a bilevel opacity value which is stored in the opacity channel.
When channeltype is "Intensity", only the color samples are dithered. When channeltype is "opacity" or "matte", only the opacity channel is dithered. When a color channel is specified, only that channel is dithered.
The choices for N are 2 through 7. The image is divided into NxN pixel tiles. In each tile, some or all pixels are turned to white depending on their intensity. For each N, (N**2)+1 levels of gray can be represented. For N == 2, 3, or 4, the pixels are turned to white in an order that maximizes dispersion (i.e., reduces granularity), while for N == 5, 6, and 7, they are turned to white in an order that creates a roughly circular black blob in the middle of each tile. An attractive "half-tone" looking image can be obtained by first rotating the image 45 degrees, performing a 5x5 ordered-dither operation, then rotating it back to the original orientation and cropping to the original image dimensions. If the original image is gamma-encoded, it is adviseable to convert it to linear intensity first, e.g., with the "-gamma 0.45455" option.
Use -output-directory to specify a directory under which to write the output files. Normally mogrify overwrites the input files, but with this option the output files may be written to a different directory tree so that the input files are preserved. The algorithm used preserves all of the input path specification in the output path so that the user-specified input path (including any sub-directory part) is appended to the output path. If the input file lacks an extension, then a suitable extension is automatically added to the output file. The user is responsible for creating the output directory specified as an argument, but subdirectories will be created as needed if the -create-directories option is supplied. This option may be used to apply transformations on files from one directory and write the transformed files to a different directory. In conjunction with -create-directories, this option is designed to support transforming whole directory trees of files provided that the relative path of the input file is included as part the list of filenames.
Sets the image orientation attribute. The image orientation attribute is compatible with the TIFF orientation tag (and the EXIF orientation tag). Accepted values are undefined, TopLeft, TopRight, BottomRight, BottomLeft, LeftTop, RightTop, RightBottom, LeftBottom, and hyphenated versions thereof (e.g. left-bottom). Please note that GraphicsMagick does not include an EXIF editor so if an EXIF profile is written to the output image, the value in the EXIF profile might not match the image. It is possible for an image file to indicate its orientation in several different ways simultaneously.
Use this option to specify the dimensions of the PostScript page in dots per inch or a TEXT page in pixels. The choices for a PostScript page are:
11x17 792 1224
Ledger 1224 792
Legal 612 1008
Letter 612 792
LetterSmall 612 792
ArchE 2592 3456
ArchD 1728 2592
ArchC 1296 1728
ArchB 864 1296
ArchA 648 864
A0 2380 3368
A1 1684 2380
A2 1190 1684
A3 842 1190
A4 595 842
A4Small 595 842
A5 421 595
A6 297 421
A7 210 297
A8 148 210
A9 105 148
A10 74 105
B0 2836 4008
B1 2004 2836
B2 1418 2004
B3 1002 1418
B4 709 1002
B5 501 709
C0 2600 3677
C1 1837 2600
C2 1298 1837
C3 918 1298
C4 649 918
C5 459 649
C6 323 459
Flsa 612 936
Flse 612 936
HalfLetter 396 612
For convenience you can specify the page size by media (e.g. A4, Ledger, etc.). Otherwise, -page behaves much like -geometry (e.g. -page letter+43+43>).
This option is also used to place subimages when writing to a multi-image format that supports offsets, such as GIF89 and MNG. When used for this purpose the offsets are always measured from the top left corner of the canvas and are not affected by the -gravity option. To position a GIF or MNG image, use -page{+-}<x>{+-}<y> (e.g. -page +100+200). When writing to a MNG file, a -page option appearing ahead of the first image in the sequence with nonzero width and height defines the width and height values that are written in the MHDR chunk. Otherwise, the MNG width and height are computed from the bounding box that contains all images in the sequence. When writing a GIF89 file, only the bounding box method is used to determine its dimensions.
For a PostScript page, the image is sized as in -geometry and positioned relative to the lower left hand corner of the page by {+-}<xoffset>{+-}<y offset>. Use -page 612x792>, for example, to center the image within the page. If the image size exceeds the PostScript page, it is reduced to fit the page. The default gravity for the -page option is NorthWest, i.e., positive x and y offset are measured rightward and downward from the top left corner of the page, unless the -gravity option is present with a value other than NorthWest.
The default page dimensions for a TEXT image is 612x792.
This option is used in concert with -density.
Use +page to remove the page settings for an image.
Each pixel is replaced by the most frequent color in a circular neighborhood whose width is specified with radius.
Pause for the specified number of seconds before repeating the animation.
Pause for the specified number of seconds before taking the next snapshot.
Use this option to disable reading the image pixels so that image characteristics such as the image dimensions may be obtained very quickly. For identify, use +ping to force reading the image pixels so that the pixel read rate may be included in the displayed information.
Use this option to affect the preview operation of an image (e.g. convert file.png -preview Gamma Preview:gamma.png). Choose from these previews:
Rotate
Shear
Roll
Hue
Saturation
Brightness
Gamma
Spiff
Dull
Grayscale
Quantize
Despeckle
ReduceNoise
AddNoise
Sharpen
Blur
Threshold
EdgeDetect
Spread
Shade
Raise
Segment
Solarize
Swirl
Implode
Wave
OilPaint
CharcoalDrawing
JPEG
The default preview is JPEG.
The command argument has the form module=arg1,arg2,arg3,...,argN where module is the name of the module to invoke (e.g. "Analyze") and arg1,arg2,arg3,...,argN are an arbitrary number of arguments to pass to the process module. The sequence of images is terminated by the appearance of any option.
If the -process option appears after all of the input images, all images are processed.
For example:
gm convert logo: -process Analyze= \
-format "%[BrightnessMean],%[BrightnessStddev]" info:-
51952,23294
add ICM, IPTC, or generic profile to image -profile filename adds an ICM (ICC color management), IPTC (newswire information), or a generic (including Exif) profile to the image Use +profile icm, +profile iptc, or +profile profile_name to remove the respective profile. Multiple profiles may be listed, separated by commas. Profiles may be excluded from subsequent listed matches by preceding their name with an exclamation point. For example, +profile '!icm,*' strips all profiles except for the ICM profile. Use identify -verbose to find out what profiles are in the image file. Use +profile "*" to remove all profiles. Writing the image to a format that does not support profiles will of course also cause all profiles to be removed. The JPEG and PNG formats will store any profiles that have been read and not removed. In JPEG they are stored in APP1 markers, and in PNG they are stored as hex-coded binary in compressed zTXt chunks, except for the iCC chunk which is stored in the iCCP chunk. To extract a profile, the -profile option is not used. Instead, simply write the file to an image format such as APP1, 8BIM, ICM, or IPTC. For example, to extract the Exif data (which is stored in JPEG files in the APP1 profile), useNote that GraphicsMagick does not attempt to update any profile to reflect changes made to the image, e.g., rotation from portrait to landscape orientation, so it is possible that the preserved profile may contain invalid data.
gm convert cockatoo.jpg exifdata.app1
Use this option to preserve the original modification and access timestamps of the file, even if it has been modified.
By default, when an image is displayed, a progress monitor bar is shown in the top left corner of an existing image display window, and the current cursor is replaced with an hourglass cursor. Use +progress to disable the progress monitor and busy cursor during display operations. While the progress monitor is disabled for all operations, the busy cursor continues to be enabled for non-display operations such as image processing. This option is useful for non-interactive display operations, or when a "clean" look is desired.
For the MIFF image format, and the TIFF format while using ZIP compression, quality/10 is the zlib compression level, which is 0 (worst but fastest compression) to 9 (best but slowest). It has no effect on the image appearance, since the compression is always lossless.
For the JPEG-2000 image format, quality is mapped using a non-linear equation to the compression ratio required by the Jasper library. This non-linear equation is intended to loosely approximate the quality provided by the JPEG v1 format. The default quality value 75 results in a request for 16:1 compression. The quality value 100 results in a request for non-lossy compression.
For the MNG and PNG image formats, the quality value sets the zlib compression level (quality / 10) and filter-type (quality % 10). Compression levels range from 0 (fastest compression) to 100 (best but slowest). For compression level 0, the Huffman-only strategy is used, which is fastest but not necessarily the worst compression.
If filter-type is 4 or less, the specified filter-type is used for all scanlines:
0: none
1: sub
2: up
3: average
4: Paeth
If filter-type is 5, adaptive filtering is used when quality is greater than 50 and the image does not have a color map, otherwise no filtering is used.
If filter-type is 6, adaptive filtering with minimum-sum-of-absolute-values is used.
Only if the output is MNG, if filter-type is 7, the LOCO color transformation and adaptive filtering with minimum-sum-of-absolute-values are used.
The default is quality is 75, which means nearly the best compression with adaptive filtering. The quality setting has no effect on the appearance of PNG and MNG images, since the compression is always lossless.
For further information, see the PNG specification.
When writing a JNG image with transparency, two quality values are required, one for the main image and one for the grayscale image that conveys the opacity channel. These are written as a single integer equal to the main image quality plus 1000 times the opacity quality. For example, if you want to use quality 75 for the main image and quality 90 to compress the opacity data, use -quality 90075.
For the PNM family of formats (PNM, PGM, and PPM) specify a quality factor of zero in order to obtain the ASCII variant of the format. Note that -compress none used to be used to trigger ASCII output but provided the opposite result of what was expected as compared with other formats.
For the TIFF format, the JPEG, WebP, Zip, and Zstd compression algorithms are influenced by the quality value. JPEG and WebP provide lossy compression so higher quality produces a larger file with less degradation. The Zip and Zstd compression algorithms (and WebP in lossless mode) are lossless and for these algorithms a higher ´quality' means to work harder to produce a smaller file, but with no difference in image quality.
This will create a 3-D effect. See -geometry for details details about the geometry specification. Offsets are not used.
Use -raise to create a raised effect, otherwise use +raise.
The channel or channels specified in the <channeltype> argument are reduced to binary, using an random-threshold method. The choices for channeltype are All, Intensity, Red, Green, Blue, Cyan, Magenta, Yellow, Black, and Opacity
When channeltype is "All", the color samples are thresholded into a graylevel and then that gray level is stored in the three color channels. Separately, the opacity channel is thresholded into a bilevel opacity value which is stored in the opacity channel. For each pixel, a new random number is used to establish the threshold to be used. The threshold never exceeds the specified maximum (HIGH) and is never less than the specified minimum (LOW).
When channeltype is "intensity", only the color samples are thresholded. When channeltype is "opacity" or "matte", only the opacity channel is thresholded. The other named channels only threshold the associated channel.
A user supplied color translation matrix (expressed as a text string) is used to translate/blend the image channels based on weightings in a supplied matrix which may be of order 3 (color channels only), 4 (color channels plus opacity), or 5 (color channels plus opacity and offset). Values in the columns of the matrix (red, green, blue, opacity) are used as multipliers with the existing channel values and added together according to the rows of the matrix. Matrix values are floating point and may be negative. The offset column (column 5) is purely additive and is scaled such that 0.0 to 1.0 represents the maximum quantum range (but values are not limited to this range). The math for the color translation matrix is similar to that used by Adobe Flash except that the offset is scaled to 1.0 (divide Flash offset by 255 for use with GraphicsMagick) so that the results are independent of quantum depth.
An identity matrix exists for each matrix order which results in no change to the image. The translation matrix should be based on an alteration of the identity matrix.
Identity matrix of order 3
1 0 0
0 1 0
0 0 1
which may be formatted into a convenient matrix argument similar to (comma is treated as white space):
-recolor "1 0 0, 0 1 0, 0 0 1"
Identity matrix of order 4
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
Identity matrix of order 5. The last row is required to exist for the purpose of parsing, but is otherwise not used.
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
As an example, an image wrongly in BGR channel order may be converted to RGB using this matrix (blue->red, red->blue):
0 0 1
0 1 0
1 0 0
and an RGB image using standard Rec.709 primaries may be converted to grayscale using this matrix of standard weighting factors:
0.2126 0.7152 0.0722
0.2126 0.7152 0.0722
0.2126 0.7152 0.0722
and contrast may be reduced by scaling down by 80% and adding a 10% offset:
0.8 0.0 0.0 0.0 0.1
0.0 0.8 0.0 0.0 0.1
0.0 0.0 0.8 0.0 0.1
0.0 0.0 0.0 0.8 0.1
0.0 0.0 0.0 0.0 1.0
The x and y offsets are treated in the same manner as in -crop.
The -remote command sends a command to a "gm display" or "gm animate" which is already running. The only command recognized at this time is the name of an image file to load. This capability is very useful to load new images without needing to restart GraphicsMagick (e.g. for a slide-show or to use GraphicsMagick as the display engine for a different GUI). Also see the +progress option for a way to disable progress indication for a clean look while loading new images.
Use +render to turn off rendering vector operations. This is useful when saving the result to vector formats such as MVG or SVG.
Adjust the current image page canvas and position based on a relative page specification. This option may be used to change the location of a subframe (e.g. part of an animation) prior to composition. If the geometry specification is absolute (includes a '!'), then the offset adjustment is absolute and there is no adjustment to page width and height, otherwise the page width and height values are also adjusted based on the current image dimensions. Use +repage to set the image page offsets to default.
Resize the image so that its rendered size remains the same as the original at the specified target resolution. Either the current image resolution units or the previously set with -units are used to interpret the argument. For example, if a 300 DPI image renders at 3 inches by 2 inches on a 300 DPI device, when the image has been resampled to 72 DPI, it will render at 3 inches by 2 inches on a 72 DPI device. Note that only a small number of image formats (e.g. JPEG, PNG, and TIFF) are capable of storing the image resolution. For formats which do not support an image resolution, the original resolution of the image must be specified via -density on the command line prior to specifying the resample resolution.
Note that Photoshop stores and obtains image resolution from a proprietary embedded profile. If this profile exists in the image, then Photoshop will continue to treat the image using its former resolution, ignoring the image resolution specified in the standard file header.
Some image formats (e.g. PNG) require use of metric or english units so even if the original image used a particular unit system, if it is saved to a different format prior to resampling, then it may be necessary to specify the desired resolution units using -units since the original units may have been lost. In other words, do not assume that the resolution units are restored if the image has been saved to a file.
This is an alias for the -geometry option and it behaves in the same manner. If the -filter option precedes the -resize option, the specified filter is used.
There are some exceptions:
When used as a composite option, -resize conveys the preferred size of the output image, while -geometry conveys the size and placement of the composite image within the main image.
When used as a montage option, -resize conveys the preferred size of the montage, while -geometry conveys information about the tiles.
See -geometry for details the geometry specification. The x and y offsets are not affected by the -gravity option.
A negative x offset rolls the image left-to-right. A negative y offset rolls the image top-to-bottom.
Positive angles rotate the image in a clockwise direction while negative angles rotate counter-clockwise.
Use > to rotate the image only if its width exceeds the height. < rotates the image only if its width is less than the height. For example, if you specify -rotate "-90>" and the image size is 480x640, the image is not rotated. However, if the image is 640x480, it is rotated by -90 degrees. If you use > or <, enclose it in quotation marks to prevent it from being misinterpreted as a file redirection.
Empty triangles left over from rotating the image are filled with the color defined as background (class backgroundColor). The color is specified using the format described under the -fill option.
See -geometry for details about the geometry specification. -sample ignores the -filter selection if the -filter option is present. Offsets, if present in the geometry string, are ignored, and the -gravity option has no effect.
This option specifies the sampling factors to be used by the DPX, JPEG, MPEG, or YUV encoders for chroma downsampling. The sampling factor must be specified while reading the raw YUV format since it is not preserved in the file header. Industry-standard video subsampling notation such as "4:2:2" may also be used to specify the sampling factors. "4:2:2" is equivalent to a specification of "2x1"
The JPEG decoder obtains the original sampling factors (and quality settings) when a JPEG file is read. To re-use the original sampling factors (and quality setting) when JPEG is output, use the -define jpeg:preserve-settings flag.
See -geometry for details about the geometry specification. -scale uses a simpler, faster algorithm, and it ignores the -filter selection if the -filter option is present. Offsets, if present in the geometry string, are ignored, and the -gravity option has no effect.
This option sets the scene number of an image or the first image in an image sequence.
Each image in the range is read with the filename followed by a period (.) and the decimal scene number. You can change this behavior by embedding a %d, %0Nd, %o, %0No, %x, or %0Nx printf format specification in the file name. For example,
gm montage -scenes 5-7 image.miff montage.miff
makes a montage of files image.miff.5, image.miff.6, and image.miff.7, and
gm animate -scenes 0-12 image%02d.miff
animates files image00.miff, image01.miff, through image12.miff.
This option indicates that the GetImage request used to obtain the image should be done on the root window, rather than directly on the specified window. In this way, you can obtain pieces of other windows that overlap the specified window, and more importantly, you can capture menus or other popups that are independent windows but appear over the specified window.
Set a named image attribute. The attribute is set on the current (previously specified on command line) image.
Unset a named image attribute. The attribute is removed from the current (previously specified on command line) image.
Segment an image by analyzing the histograms of the color components and identifying units that are homogeneous with the fuzzy c-means technique.
Segmentation is a very useful fast and and approximate color quantization algorithm for scanned printed pages or scanned cartoons. It may also be used as a special effect. Specify cluster threshold as the minimum percentage of total pixels in a cluster before it is considered valid. For huge images containing small detail, this may need to be a tiny fraction of a percent (e.g. 0.015) so that important detail is not lost. Smoothing threshold eliminates noise in the second derivative of the histogram. As the value is increased, you can expect a smoother second derivative. The default is 1.5. Add the -verbose option to see a dump of cluster statistics given the parameters used. The statistics may be used as a guide to help fine tune the options.
Specify azimuth and elevation as the position of the light source. Use +shade to return the shading results as a grayscale image.
This option specifies whether the utility should attempt to use shared memory for pixmaps. GraphicsMagick must be compiled with shared memory support, and the display must support the MIT-SHM extension. Otherwise, this option is ignored. The default is True.
Use a Gaussian operator of the given radius and standard deviation (sigma).
Specify the width of the region to be removed from both sides of the image and the height of the regions to be removed from top and bottom.
Use the specified positive or negative shear angle.
Shearing slides one edge of an image along the X or Y axis, creating a parallelogram. An X direction shear slides an edge along the X axis, while a Y direction shear slides an edge along the Y axis. The amount of the shear is controlled by a shear angle. For X direction shears, x degrees is measured relative to the Y axis, and similarly, for Y direction shears y degrees is measured relative to the X axis.
Empty triangles left over from shearing the image are filled with the color defined as background (class backgroundColor). The color is specified using the format described under the -fill option.
Use this option to specify the width and height of raw images whose dimensions are unknown such as GRAY, RGB, or CMYK. In addition to width and height, use -size with an offset to skip any header information in the image or tell the number of colors in a MAP image file, (e.g. -size 640x512+256).
For Photo CD images, choose from these sizes:
192x128
384x256
768x512
1536x1024
3072x2048
Finally, use this option to choose a particular resolution layer of a JBIG or JPEG image (e.g. -size 1024x768).
Use this option to grab more than one image from the X server screen, to create an animation sequence.
Specify factor as the percent threshold of the intensity (0 - 99.9%).
This option produces a solarization effect seen when exposing a photographic film to light during the development process.
Amount defines the size of the neighborhood around each pixel to choose a candidate pixel to swap.
Use an offset to start the image hiding some number of pixels from the beginning of the image. Note this offset and the image size. You will need this information to recover the steganographic image (e.g. display -size 320x256+35 stegano:image.png).
The left side of the stereo pair is saved as the red channel of the output image. The right side is saved as the green channel. Red-green stereo glasses are required to properly view the stereo image.
All embedded profiles and text attributes are stripped from the image. This is useful for images used for the web, or when output files need to be as small as possible
Be careful not to use this option to remove author, copyright, and license information that you are required to retain when redistributing an image.
The color is specified using the format described under the -fill option.
See -draw for further details.
See -draw for further details.
Degrees defines the tightness of the swirl.
Specifies the name of the preferred font to use in fixed (typewriter style) formatted text. The default is 14 point Courier.
You can tag a font to specify whether it is a PostScript, TrueType, or X11 font. For example, Courier.ttf is a TrueType font and x:fixed is X11.
Modify the image such that any pixel sample with an intensity value greater than the threshold is assigned the maximum intensity (white), or otherwise is assigned the minimum intensity (black). If a percent prefix is applied, then the threshold is a percentage of the available range.
To efficiently create a black and white image from a color image, use
gm convert -threshold 50% in.png out.png
The optimum threshold value depends on the nature of the image. In order to threshold individual channels, use the -operator subcommand with it's Threshold, Threshold-White, or Threshold-Black options.
The -thumbnail command resizes the image as quickly as possible, with more concern for speed than resulting image quality. Regardless, resulting image quality should be acceptable for many uses. It is primarily intended to be used to generate smaller versions of the image, but may also be used to enlarge the image. The -thumbnail geometry argument observes the same syntax and rules as it does for -resize.
Use this option to assign a specific title to the image. This is assigned to the image window and is typically displayed in the window title bar. Optionally you can include the image filename, type, width, height, Exif data, or other image attribute by embedding special format characters described under the -format option.
For example,
-title "%m:%f %wx%h"
produces an image title of MIFF:bird.miff 512x480 for an image titled bird.miff and whose width is 512 and height is 480.
This option applies the transformation matrix from a previous -affine option.
gm convert -affine 2,2,-2,2,0,0 -transform bird.ppm bird.jpg
The color is specified using the format described under the -fill option.
Normally, this integer value is zero or one. A value of zero or one causes the use of an optimal tree depth for the color reduction algorithm
An optimal depth generally allows the best representation of the source image with the fastest computational speed and the least amount of memory. However, the default depth is inappropriate for some images. To assure the best representation, try values between 2 and 8 for this parameter. Refer to quantize for more details.
The -colors or -monochrome option, or writing to an image format which requires color reduction, is required for this option to take effect.
This option removes any edges that are exactly the same color as the corner pixels. Use -fuzz to make -trim remove edges that are nearly the same color as the corner pixels.
Choose from: Bilevel, Grayscale, Palette, PaletteMatte, TrueColor, TrueColorMatte, ColorSeparation, ColorSeparationMatte, or Optimize.
Normally, when a format supports different subformats such as bilevel, grayscale, palette, truecolor, and truecolor+alpha, the encoder will try to choose a suitable subformat based on the nature of the image. The -type option may be used to tailor the output subformat. By default the output subformat is based on readily available image information and is usually similar to the input format.
Specify -type Optimize in order to enable inspecting all pixels (if necessary) in order to find the most efficient subformat. Inspecting all of the pixels may be slow for very large images, particularly if they are stored in a disk cache. If an RGB image contains only gray pixels, then every pixel in the image must be inspected in order to decide that the image is actually grayscale!
Sometimes a specific subformat is desired. For example, to force a JPEG image to be written in TrueColor RGB format even though only gray pixels are present, use
gm convert bird.pgm -type TrueColor bird.jpg
Similarly, using -type TrueColorMatte will force the encoder to write an alpha channel even though the image is opaque, if the output format supports transparency.
Some pseudo-formats (e.g. the XC format) will respect the requested type if it occurs previously on the command line. For example, to obtain a DirectClass solid color canvas image rather than PsuedoClass, use
gm convert -size 640x480 -type TrueColor xc:red red.miff
Likewise, specify -type Bilevel, Grayscale, TrueColor, or TrueColorMatte prior to reading a Postscript (or PDF file) in order to influence the type of image that Ghostcript returns. Reading performance will be dramatically improved for black/white Postscript if Bilevel is specified, and will be considerably faster if Grayscale is specified.
Suppose that while you are displaying an image the file that is currently displayed is over-written. display will automatically detect that the input file has been changed and update the displayed image accordingly.
Choose from: Undefined, PixelsPerInch, or PixelsPerCentimeter. This option is normally used in conjunction with the -density option.
The -unsharp option sharpens an image. The image is convolved with a Gaussian operator of the given radius and standard deviation (sigma). For reasonable results, radius should be larger than sigma. Use a radius of 0 to have the method select a suitable radius.
The parameters are:
radius
The radius of the Gaussian, in pixels, not counting the center pixel (default 0).
sigma
The standard deviation of the Gaussian, in pixels (default 1.0).
amount
The percentage of the difference between the original and the blur image that is added back into the original (default 1.0).
threshold
The threshold, as a fraction of MaxRGB, needed to apply the difference amount (default 0.05).
This information is printed: image scene number; image name; image size; the image class (DirectClass or PseudoClass); the total number of unique colors; and the number of seconds to read and transform the image. If the image is DirectClass, the total number of unique colors is not displayed unless -verbose is specified twice since it may take quite a long time to compute, particularly for deep images. If the image is PseudoClass then its pixels are defined by indexes into a colormap. If the image is DirectClass then each pixel includes a complete and independent color specification.
If -colors is also specified, the total unique colors in the image and color reduction error values are printed. Refer to quantize for a description of these values.
This option defines "virtual pixels" for use in operations that can access pixels outside the boundaries of an image.
Choose from these methods:
Constant
Use the image background color.
Edge
Extend the edge pixel toward infinity (default).
Mirror
Mirror the image.
Tile
Tile the image.
This option affects operations that use virtual pixels such as -blur, -sharpen, -wave, etc.
Choose from these visual classes:
StaticGray
GrayScale
StaticColor
PseudoColor
TrueColor
DirectColor
default
visual id
The X server must support the visual you choose, otherwise an error occurs. If a visual is not specified, the visual class that can display the most simultaneous colors on the default screen is chosen.
Specify amplitude and wavelength of the wave.
Use -white-threshold to set pixels with values above the specified threshold to maximum value (white). If only one value is supplied, or the red, green, and blue values are identical, then intensity thresholding is used. If the color threshold values are not identical then channel-based thresholding is used, and color distortion will occur. Specify a negative value (e.g. -1) if you want a channel to be ignored but you do want to threshold a channel later in the list. If a percent (%) symbol is appended, then the values are treated as a percentage of maximum range.
id can be a window id or name. Specify root to select X's root window as the target window.
By default the image is tiled onto the background of the target window. If backdrop or -geometry are specified, the image is surrounded by the background color. Refer to X RESOURCES for details.
The image will not display on the root window if the image has more unique colors than the target window colormap allows. Use -colors to reduce the number of colors.
The current image is written to the specified filename and then processing continues using that image. The following is an example of how several sizes of an image may be generated in one command (repeat as often as needed):
gm convert input.jpg -resize 50% -write input50.jpg \
-resize 25% input25.jpg
If filename already exists, you will be prompted as to whether it should be overwritten.
By default, the image is written in the format that it was read in as. To specify a particular image format, prefix filename with the image type and a colon (e.g., ps:image) or specify the image type as the filename suffix (e.g., image.ps). Specify file as - for standard output. If file has the extension .Z or .gz, the file size is compressed using compress or gzip respectively. Precede the image file name with | to pipe to a system command.
Use -compress to specify the type of image compression.
The equivalent X resource for this option is writeFilename (class WriteFilename). See "X Resources", below, for details.
GraphicsMagick uses a number of XML format configuration files:
<?xml version="1.0"?>
<colormap>
<color name="AliceBlue" red="240" green="248" blue="255"
compliance="SVG, X11, XPM" />
</colormap>
<?xml version="1.0"?>
<magicklog>
<log events="None" />
<log output="stdout" />
<log filename="Magick-%d.log" />
<log generations="3" />
<log limit="2000" />
<log format="%t %r %u %p %m/%f/%l/%d:\n %e" />
</magicklog>
<?xml version="1.0"?>
<modulemap>
<module magick="8BIM" name="META" />
</modulemap>
<?xml version="1.0"?>
<typemap>
<include file="type-windows.mgk" />
<type
name="AvantGarde-Book"
fullname="AvantGarde Book"
family="AvantGarde"
foundry="URW"
weight="400"
style="normal"
stretch="normal"
format="type1"
metrics="/usr/local/share/ghostscript/fonts/a010013l.afm"
glyphs="/usr/local/share/ghostscript/fonts/a010013l.pfb"
/>
</typemap>
Animate displays a sequence of images on any workstation display running an X server. animate first determines the hardware capabilities of the workstation. If the number of unique colors in an image is less than or equal to the number the workstation can support, the image is displayed in an X window. Otherwise the number of colors in the image is first reduced to match the color resolution of the workstation before it is displayed.
This means that a continuous-tone 24 bits-per-pixel image can display on a 8 bit pseudo-color device or monochrome device. In most instances the reduced color image closely resembles the original. Alternatively, a monochrome or pseudo-color image sequence can display on a continuous-tone 24 bits-per-pixel device.
To help prevent color flashing on X server visuals that have colormaps, animate creates a single colormap from the image sequence. This can be rather time consuming. You can speed this operation up by reducing the colors in the image before you "animate" them. Use mogrify to color reduce the images to a single colormap. See mogrify(1) for details. Alternatively, you can use a Standard Colormap; or a static, direct, or true color visual. You can define a Standard Colormap with xstdcmap. See xstdcmap(1) for details. This method is recommended for colormapped X server because it eliminates the need to compute a global colormap.
To animate a set of images of a cockatoo, use:
gm animate cockatoo.*
To animate a cockatoo image sequence while using the Standard Colormap best, use:
xstdcmap -best
gm animate -map best cockatoo.*
To animate an image of a cockatoo without a border centered on a backdrop, use:
gm animate +borderwidth -backdrop cockatoo.*
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
Any option you specify on the command line remains in effect for the group of images following it, until the group is terminated by the appearance of any option or -noop. For example, to animate three images, the first with 32 colors, the second with an unlimited number of colors, and the third with only 16 colors, use:
gm animate -colors 32 cockatoo.1 -noop cockatoo.2
-colors 16 cockatoo.3
Animate options can appear on the command line or in your X resources file. See X(1). Options on the command line supersede values specified in your X resources file. Image filenames may appear in any order on the command line if the image format is MIFF (refer to miff(5) and the scene keyword is specified in the image. Otherwise the images will display in the order they appear on the command line.
Press any button to map or unmap the Command widget. See the next section for more information about the Command widget.
The Command widget lists a number of sub-menus and commands. They are
Animate
Open
Play
Step
Repeat
Auto Reverse
Speed
Faster
Slower
Direction
Forward
Reverse
Image Info
Help
Quit
Menu items with a indented triangle have a sub-menu. They are represented above as the indented items. To access a sub-menu item, move the pointer to the appropriate menu and press a button and drag. When you find the desired sub-menu item, release the button and the command is executed. Move the pointer away from the sub-menu if you decide not to execute a particular command.
Ctl+O
Press to load an image from a file.
space
Press to display the next image in the sequence.
<
Press to speed-up the display of the images. Refer to -delay for more
information.
>
Press to slow the display of the images. Refer to -delay for more
information.
?
Press to display information about the image. Press any key or button to erase
the information.
This information is printed: image name; image size; and the total number of
unique colors in the image.
F1
Press to display helpful information about animate(1).
Ctl-q
Press to discard all images and exit program.
Animate options can appear on the command line or in your X resource file. Options on the command line supersede values specified in your X resource file. See X(1) for more information on X resources.
All animate options have a corresponding X resource. In
addition, the animate program uses the following X resources:
background (class Background)
Specifies the preferred color to use for the Image window
background. The default is #ccc.
borderColor (class BorderColor)
Specifies the preferred color to use for the Image window border.
The default is #ccc.
borderWidth (class BorderWidth)
Specifies the width in pixels of the Image window border. The
default is 2.
font (class Font or FontList)
Specifies the name of the preferred font to use in normal
formatted text. The default is 14 point Helvetica.
foreground (class Foreground)
Specifies the preferred color to use for text within the Image
window. The default is black.
geometry (class geometry)
Specifies the preferred size and position of the image window. It
is not necessarily obeyed by all window managers. Offsets, if present, are
handled in X(1) style. A negative x offset is measured from the right
edge of the screen to the right edge of the icon, and a negative y offset is
measured from the bottom edge of the screen to the bottom edge of the icon.
iconGeometry (class IconGeometry)
Specifies the preferred size and position of the application when
iconified. It is not necessarily obeyed by all window managers. Offsets, if
present, are handled in the same manner as in class Geometry.
iconic (class Iconic)
This resource indicates that you would prefer that the
application's windows initially not be visible as if the windows had be
immediately iconified by you. Window managers may choose not to honor the
application's request.
matteColor (class MatteColor)
Specify the color of windows. It is used for the backgrounds of
windows, menus, and notices. A 3D effect is achieved by using highlight and
shadow colors derived from this color. Default value: #ddd.
name (class Name)
This resource specifies the name under which resources for the
application should be found. This resource is useful in shell aliases to
distinguish between invocations of an application, without resorting to
creating links to alter the executable file name. The default is the
application name.
sharedMemory (class SharedMemory)
This resource specifies whether animate should attempt use shared
memory for pixmaps. ImageMagick must be compiled with shared memory support,
and the display must support the MIT-SHM extension. Otherwise, this resource
is ignored. The default is True.
text_font (class textFont)
Specifies the name of the preferred font to use in fixed
(typewriter style) formatted text. The default is 14 point Courier.
title (class Title)
This resource specifies the title to be used for the Image window.
This information is sometimes used by a window manager to provide some sort
of header identifying the window. The default is the image file name.
batch executes an arbitary number of the utility commands (e.g. convert) in the form of a simple linear batch script in order to improve execution efficiency, and/or to allow use as a subordinate co-process under the control of an arbitrary script or program.
To drive 'gm batch' using a shell script (or a program written in any language), have the script/program send commands to 'gm batch' via its standard input. Specify that standard input should be used by using '-' as the file name. The following example converts all files matching '*.jpg' to TIFF format while rotating each file by 90 degrees and stripping all embedded profiles. The shell script syntax is standard Unix shell:
for file in *.jpg
do
outfile=`basename $file .jpg`.tiff
echo convert -verbose "'$file'" -rotate 90 \
+profile "'*'" "'$outfile'"
done | gm batch -echo on -feedback on -
We can accomplish the same as the previous example by putting all the commands in a text file and then specifying the name of the text file as the script to execute:
for file in *.jpg
do
outfile=`basename $file .jpg`.tiff
echo convert -verbose "'$file'" -rotate 90 \
+profile "'*'" "'$outfile'"
done > script.txt
gm batch -echo on -feedback on script.txt
Options are processed from left to right and must appear before any filename argument.
Specify on to enable echoing commands to standard output as they are read or off to disable. The default is off.
Commands must be parsed from the input stream and escaping needs to be used to protect spaces or quoting characters in the input. Specify unix to use unix-style command line parsing or windows for Microsoft Windows command shell style parsing. The default depends on if the software is compiled for Microsoft Windows or for a Unix-type system (including Cygwin on Microsoft Windows). It is recommended to use unix syntax because it is more powerful and more portable.
When feedback is enabled, this specifies the text to print when the command fails. The default text is FAIL.
Print text (see -pass and -fail options) feedback after each command to indicate the result, the default is off.
Prints batch command help.
When feedback is enabled, this specifies the text to print when the command passes. The default text is PASS.
If no filename argument was specified, a simple command prompt appears where you may enter GraphicsMagick commands. The default prompt is GM>. Use this option to change the prompt to something else.
Normally command processing continues if a command encounters an error. Specify -stop-on-error on to cause processing to quit immediately on error.
benchmark executes an arbitrary gm utility command (e.g. convert) for one or more loops, and/or a specified execution time, and reports many execution metrics. For builds using OpenMP, a mode is provided to execute the benchmark with an increasing number of threads and provide a report of speedup and multi-thread execution efficiency. If benchmark is used to execute a command without any additional benchmark options, then the command is run once.
To obtain benchmark information for a single execution of a command:
To obtain benchmark information from 100 iterations of the command:
gm benchmark convert input.ppm -gaussian 0x1 output.ppm
To obtain benchmark information by iterating the command until a specified amount of time (in seconds) has been consumed:
gm benchmark -iterations 100 convert input.ppm \
-gaussian 0x1 output.ppm
To obtain a full performance report with an increasing number of threads (1-32 threads, stepping the number of threads by four each time):
gm benchmark -duration 30 convert input.ppm \
-gaussian 0x1 output.ppm
Here is the interpretation of the output:
gm benchmark -duration 3 -stepthreads 4 convert \
input.ppm -gaussian 0x2 output.ppm
threads - number of threads used.
iter - number of command iterations executed.
user - total user time consumed.
total - total elapsed time consumed.
iter/s - number of command iterations per second.
iter/cpu - amount of CPU time consumed per iteration.
speedup - speedup compared with one thread.
karp-flatt - Karp-Flatt measure of speedup efficiency.
Please note that the reported "speedup" is based on the execution time of just one thread. A preliminary warm-up pass is used before timing the first loop in order to ensure that the CPU is brought out of power-saving modes and that system caches are warmed up. Most modern CPUs provide a "turbo" mode where the CPU clock speed is increased (e.g. by a factor of two) when only one or two cores are active. If the CPU grows excessively hot (due to insufficient cooling), then it may dial back its clock rates as a form of thermal management. These factors result in an under-reporting of speedup compared to if "turbo" mode was disabled and the CPU does not need to worry about thermal management. The powertop utility available under Linux and Solaris provides a way to observe CPU core clock rates while a benchmark is running.
Options are processed from left to right and must appear before any argument.
Prints benchmark command help.
compare compares two similar images using a specified statistical method (see -metric) and/or by writing a difference image (-file), with the altered pixels annotated using a specified method (see -highlight-style) and color (see -highlight-color). Reference-image is the original image and compare-image is the (possibly) altered version, which should have the same dimensions as reference-image.
To compare two images using Mean Square Error (MSE) statistical analysis use:
gm compare -metric mse original.miff compare.miff
To create an annotated difference image use:
gm compare -highlight-style assign -highlight-color purple \
-file diff.miff original.miff compare.miff
Options are processed in command line order. Any option you specify on the command line remains in effect only for the image that follows. All options are reset to their default values after each image is read.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
composite composites (combines) images to create new images. base-image is the base image and change-image contains the changes. ouput-image is the result, and normally has the same dimensions as base-image.
The optional mask-image can be used to provide opacity information for change-image when it has none or if you want a different mask. A mask image is typically grayscale and the same size as base-image. If mask-image is not grayscale, it is converted to grayscale and the resulting intensities are used as opacity information.
To composite an image of a cockatoo with a perch, use:
gm composite cockatoo.miff perch.ras composite.miff
To compute the difference between images in a series, use:
gm composite -compose difference series.2 series.1
difference.miff
To composite an image of a cockatoo with a perch starting at location (100,150), use:
gm composite -geometry +100+150 cockatoo.miff
perch.ras composite.miff
To tile a logo across your image of a cockatoo, use
gm convert +shade 30x60 cockatoo.miff mask.miff
gm composite -compose bumpmap -tile logo.png
cockatoo.miff mask.miff composite.miff
To composite a red, green, and blue color plane into a single composite image, try
gm composite -compose CopyGreen green.png red.png
red-green.png
gm composite -compose CopyBlue blue.png red-green.png
gm composite.png
Options are processed in command line order. Any option you specify on the command line remains in effect only for the image that follows. All options are reset to their default values after each image is read.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
The Magick scripting language (MSL) will primarily benefit those that want to accomplish custom image processing tasks but do not wish to program, or those that do not have access to a Perl interpreter or a compiler. The interpreter is called conjure and here is an example script:
<?xml version="1.0" encoding="UTF-8"?>
<image size="400x400" >
<read filename="image.gif" />
<get width="base-width" height="base-height" />
<resize geometry="%[dimensions]" />
<get width="width" height="height" />
<print output=
"Image sized from %[base-width]x%[base-height]
to %[width]x%[height].\n" />
<write filename="image.png" />
</image>
invoked with
gm conjure -dimensions 400x400 incantation.msl
All operations will closely follow the key/value pairs defined in PerlMagick, unless otherwise noted.
Options are processed in command line order. Any option you specify on the command line remains in effect until it is explicitly changed by specifying the option again with a different effect, or if it is changed by a statement in the scripting language.
You can define your own keyword/value pairs on the command line. The script can then use this information when setting values by including %[keyword] in the string. For example, if you included "-dimensions 400x400" on the command line, as illustrated above, then any string containing "%[dimensions]" would have 400x400 substituted. The "%[string]" can be used either an entire string, such as geometry="%[dimensions]" or as a part of a string such as filename="%[basename].png".
The keyword can be any string except for the following reserved strings (in any upper, lower, or mixed case variant): debug, help, and verbose, whose usage is described below.
The value can be any string. If either the keyword or the value contains white space or any symbols that have special meanings to your shell such as "#", "|", or "%", enclose the string in quotation marks or use "\" to escape the white space and special symbols.
Keywords and values are case dependent. "Key", "key", and "KEY" would be three different keywords.
For a more detailed description of each option, see Options, above.
The Magick Scripting Language (MSL) presently defines the
following elements and their attributes:
<image>
background, color, id, size
Define a new image object. </image> destroys it. Because of this, if you wish to reference multiple "subimages" (aka pages or layers), you can embed one image element inside of another. For example:
<image>
<read filename="input.png" />
<get width="base-width" height="base-height" />
<image height="base-height" width="base-width">
<image />
<write filename="output.mng" />
</image>
<image size="400x400" />
<group>
Define a new group of image objects. By default, images are only valid for the life of their <image>element.
<image> -- creates the image
..... -- do stuff with it
</image> -- dispose of the image
However, in a group, all images in that group will stay around for the life of the group:
<group> -- start a group
<image> -- create an image
.... -- do stuff
</image> -- NOOP
<image> -- create another image
.... -- do more stuff
</image> -- NOOP
<write filename="image.mng" /> -- output
</group> -- dispose of both images
<read>
filename
Read a new image from a disk file.
<read filename="image.gif" />
To read two images use
<read filename="image.gif" />
<read filename="image.png />
<write>
filename
Write the image(s) to disk, either as a single multiple-image file or multiple
ones if necessary.
<write filename=image.tiff" />
<get width="base-width" height="base-height" />
<print output="Image size is %[base-width]x%[base-height].\n" />
<set>
background, bordercolor, clip-mask, colorspace, density, magick, mattecolor,
opacity. Set an attribute recognized by PerlMagick's GetAttribute().
<profile>
[profilename]
Read one or more IPTC, ICC or generic profiles from file and assign to image
<profile iptc="profile.iptc" generic="generic.dat" />
To remove a specified profile use "!" as the filename eg
<profile icm="!" iptc="profile.iptc" />
<border>
fill, geometry, height, width
<blur>
radius, sigma
<charcoal>
radius, sigma
<chop>
geometry, height, width, x, y
<crop>
geometry, height, width, x, y
<composite>
compose, geometry, gravity, image, x, y
<?xml version="1.0" encoding="UTF-8"?>
<group>
<image id="image_01">
<read filename="cloud3.gif"/>
<resize geometry="250x90"/>
</image>
<image id="image_02">
<read filename="cloud4.gif"/>
<resize geometry="190x100"/>
</image>
<image>
<read filename="background.jpg"/>
<composite image="image_01" geometry="+740+470"/>
<composite image="image_02" geometry="+390+415"/>
</image>
<write filename="result.png"/>
</group>
<despeckle>
<emboss>
radius, sigma
<enhance>
<equalize>
<edge>
radius
<flip>
<flop>
<frame>
fill, geometry, height, width, x, y, inner, outer
<flatten>
<get>
height, width
<gamma>
red, green, blue
<image>
background, color, id, size
<implode>
amount
<magnify>
<minify>
<medianfilter>
radius
<normalize>
<oilpaint>
radius
<print>
output
<profile>
[profilename]
<read>
<resize>
blur, filter, geometry, height, width
<roll>
geometry, x, y
<rotate>
degrees
<reducenoise>
radius
<sample>
geometry, height, width
<scale>
geometry, height, width
<sharpen>
radius, sigma
<shave>
geometry, height, width
<shear>
x, y
<solarize>
threshold
<spread>
radius
<stegano>
image
<stereo>
image
<swirl>
degrees
<texture>
image
<threshold>
threshold
<transparent>
color
<trim>
Convert converts an input file using one image format to an output file with a differing image format. In addition, various types of image processing can be performed on the converted image during the conversion process. Convert recognizes the image formats listed in GraphicsMagick(1).
To make a thumbnail of a JPEG image, use:
gm convert -size 120x120 cockatoo.jpg -resize 120x120
+profile "*" thumbnail.jpg
In this example, '-size 120x120' gives a hint to the JPEG decoder that the image is going to be downscaled to 120x120, allowing it to run faster by avoiding returning full-resolution images to GraphicsMagick for the subsequent resizing operation. The ´-resize 120x120' specifies the desired dimensions of the output image. It will be scaled so its largest dimension is 120 pixels. The ´+profile "*"' removes any ICM, EXIF, IPTC, or other profiles that might be present in the input and aren't needed in the thumbnail.
To convert a MIFF image of a cockatoo to a SUN raster image, use:
gm convert cockatoo.miff sun:cockatoo.ras
To convert a multi-page PostScript document to individual FAX pages, use:
gm convert -monochrome document.ps fax:page
To convert a TIFF image to a PostScript A4 page with the image in the lower left-hand corner, use:
gm convert -page A4+0+0 image.tiff document.ps
To convert a raw Gray image with a 128 byte header to a portable graymap, use:
gm convert -depth 8 -size 768x512+128 gray:raw
image.pgm
In this example, "raw" is the input file. Its format is "gray" and it has the dimensions and number of header bytes specified by the -size option and the sample depth specified by the -depth option. The output file is "image.pgm". The suffix ".pgm" specifies its format.
To convert a Photo CD image to a TIFF image, use:
gm convert -size 1536x1024 img0009.pcd image.tiff
gm convert img0009.pcd[4] image.tiff
To create a visual image directory of all your JPEG images, use:
gm convert 'vid:*.jpg' directory.miff
To annotate an image with blue text using font 12x24 at position (100,100), use:
gm convert -font helvetica -fill blue
-draw "text 100,100 Cockatoo"
bird.jpg bird.miff
To tile a 640x480 image with a JPEG texture with bumps use:
gm convert -size 640x480 tile:bumps.jpg tiled.png
To surround an icon with an ornamental border to use with Mosaic(1), use:
gm convert -mattecolor "#697B8F" -frame 6x6 bird.jpg
icon.png
To create a MNG animation from a DNA molecule sequence, use:
gm convert -delay 20 dna.* dna.mng
Options are processed in command line order. Any option you specify on the command line remains in effect for the set of images that follows, until the set is terminated by the appearance of any option or -noop. Some options only affect the decoding of images and others only the encoding. The latter can appear after the final group of input images.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
Display is a machine architecture independent image processing and display program. It can display an image on any workstation screen running an X server. Display can read and write many of the more popular image formats (e.g. JPEG, TIFF, PNM, Photo CD, etc.).
With display, you can perform these functions on an image:
o load an image from a file
o display the next image
o display the former image
o display a sequence of images as a slide show
o write the image to a file
o print the image to a PostScript printer
o delete the image file
o create a Visual Image Directory
o select the image to display by its thumbnail rather than name
o undo last image transformation
o copy a region of the image
o paste a region to the image
o restore the image to its original size
o refresh the image
o half the image size
o double the image size
o resize the image
o crop the image
o cut the image
o flop image in the horizontal direction
o flip image in the vertical direction
o rotate the image 90 degrees clockwise
o rotate the image 90 degrees counter-clockwise
o rotate the image
o shear the image
o roll the image
o trim the image edges
o invert the colors of the image
o vary the color brightness
o vary the color saturation
o vary the image hue
o gamma correct the image
o sharpen the image contrast
o dull the image contrast
o perform histogram equalization on the image
o perform histogram normalization on the image
o negate the image colors
o convert the image to grayscale
o set the maximum number of unique colors in the image
o reduce the speckles within an image
o eliminate peak noise from an image
o detect edges within the image
o emboss an image
o segment the image by color
o simulate an oil painting
o simulate a charcoal drawing
o annotate the image with text
o draw on the image
o edit an image pixel color
o edit the image matte information
o composite an image with another
o add a border to the image
o surround image with an ornamental border
o apply image processing techniques to a region of interest
o display information about the image
o zoom a portion of the image
o show a histogram of the image
o display image to background of a window
o set user preferences
o display information about this program
o discard all images and exit program
o change the level of magnification
o display images specified by a World Wide Web (WWW) uniform resource locator
(URL)
To scale an image of a cockatoo to exactly 640 pixels in width and 480 pixels in height and position the window at location (200,200), use:
gm display -geometry 640x480+200+200! cockatoo.miff
To display an image of a cockatoo without a border centered on a backdrop, use:
gm display +borderwidth -backdrop cockatoo.miff
To tile a slate texture onto the root window, use:
gm display -size 1280x1024 -window root slate.png
To display a visual image directory of all your JPEG images, use:
gm display 'vid:*.jpg'
To display a MAP image that is 640 pixels in width and 480 pixels in height with 256 colors, use:
gm display -size 640x480+256 cockatoo.map
To display an image of a cockatoo specified with a World Wide Web (WWW) uniform resource locator (URL), use:
gm display ftp://wizards.dupont.com/images/cockatoo.jpg
To display histogram of an image, use:
gm gm convert file.jpg HISTOGRAM:- | gm display -
Options are processed in command line order. Any option you specify on the command line remains in effect until it is explicitly changed by specifying the option again with a different effect. For example to display three images, the first with 32 colors, the second with an unlimited number of colors, and the third with only 16 colors, use:
gm display -colors 32 cockatoo.miff -noop duck.miff
-colors 16 macaw.miff
Display options can appear on the command line or in your X resources file. See X(1). Options on the command line supersede values specified in your X resources file.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
The effects of each button press is described below. Three buttons are required. If you have a two button mouse, button 1 and 3 are returned. Press ALT and button 3 to simulate button 2.
Open
Next
Former
Delete
Update
If you choose Open, the image represented by the tile is displayed. To return to the visual image directory, choose Next from the Command widget (refer to Command Widget). Next and Former moves to the next or former image respectively. Choose Delete to delete a particular image tile. Finally, choose Update to synchronize all the image tiles with their respective images. See montage and miff for more details.
The Command widget lists a number of sub-menus and commands. They are
File
Open...
Next
Former
Select...
Save...
Print...
Delete...
Canvas...
Visual Directory...
Quit
Edit
Undo
Redo
Cut
Copy
Paste
View
Half Size
Original Size
Double Size
Resize...
Apply
Refresh
Restore
Transform
Crop
Chop
Flop
Flip
Rotate Right
Rotate Left
Rotate...
Shear...
Roll...
Trim Edges
Enhance
Hue...
Saturation...
Brightness...
Gamma...
Spiff...
Dull
Equalize
Normalize
Negate
GRAYscale
Quantize...
Effects
Despeckle
Emboss
Reduce Noise
Add Noise
Sharpen...
Blur...
Threshold...
Edge Detect...
Spread...
Shade...
Raise...
Segment...
F/X
Solarize...
Swirl...
Implode...
Wave...
Oil Paint...
Charcoal Draw...
Image Edit
Annotate...
Draw...
Color...
Matte...
Composite...
Add Border...
Add Frame...
Comment...
Launch...
Region of Interest...
Miscellany
Image Info
Zoom Image
Show Preview...
Show Histogram
Show Matte
Background...
Slide Show
Preferences...
Help
Overview
Browse Documentation
About Display
Menu items with a indented triangle have a sub-menu. They are represented above as the indented items. To access a sub-menu item, move the pointer to the appropriate menu and press button 1 and drag. When you find the desired sub-menu item, release the button and the command is executed. Move the pointer away from the sub-menu if you decide not to execute a particular command.
Accelerators are one or two key presses that effect a particular command. The keyboard accelerators that display understands is:
Ctl+O Press to load an image from a file.
space Press to display the next image.
If the image is a multi-paged document such as a PostScript document, you can skip ahead several pages by preceding this command with a number. For example to display the fourth page beyond the current page, press 4space.
backspace Press to display the former image.
If the image is a multi-paged document such as a PostScript document, you can skip behind several pages by preceding this command with a number. For example to display the fourth page preceding the current page, press 4n.
Ctl-S Press to save the image to a file.
Ctl-P Press to print the image to a
PostScript printer.
Ctl-D Press to delete an image file.
Ctl-N Press to create a blank canvas.
Ctl-Q Press to discard all images and exit program.
Ctl+Z Press to undo last image transformation.
Ctl+R Press to redo last image transformation.
Ctl-X Press to cut a region of
the image.
Ctl-C Press to copy a region of
the image.
Ctl-V Press to paste a region to
the image.
< Press to halve the image size.
. Press to return to the original image size.
> Press to double the image size.
% Press to resize the image to a width and height
you specify.
Cmd-A Press to make any image transformations
permanent.
By default, any image size transformations are
applied to the original image to create the
image displayed on the X server. However, the
transformations are not permanent (i.e. the
original image does not change size only the
X image does). For example, if you press ">"
the X image will appear to double in size, but
the original image will in fact remain the same
size. To force the original image to double in
size, press ">" followed by "Cmd-A".
@ Press to refresh the image window.
C Press to crop the image.
[ Press to chop the image.
H Press to flop image in the horizontal direction.
V Press to flip image in the vertical direction.
/ Press to rotate the image 90 degrees clockwise.
\ Press to rotate the image 90 degrees
counter-clockwise.
* Press to rotate the image
the number of degrees you specify.
S Press to shear the image the number of degrees
you specify.
R Press to roll the image.
T Press to trim the image edges.
Shft-H Press to vary the color hue.
Shft-S Press to vary the color saturation.
Shft-L Press to vary the image brightness.
Shft-G Press to gamma correct the image.
Shft-C Press to spiff up the image contrast.
Shft-Z Press to dull the image contrast.
= Press to perform histogram equalization on
the image.
Shft-N Press to perform histogram normalization on
the image.
Shft-~ Press to negate the colors of the image.
. Press to convert the image colors to gray.
Shft-# Press to set the maximum number of unique
colors in the image.
F2 Press to reduce the speckles in an image.
F2 Press to emboss an image.
F4 Press to eliminate peak noise from an image.
F5 Press to add noise to an image.
F6 Press to sharpen an image.
F7 Press to blur image an image.
F8 Press to threshold the image.
F9 Press to detect edges within an image.
F10 Press to displace pixels by a random amount.
F11 Press to shade the image using a distant light
source.
F12 Press to lighten or darken image edges to
create a 3-D effect.
F13 Press to segment the image by color.
Meta-S Press to swirl image pixels about the center.
Meta-I Press to implode image pixels about the center.
Meta-W Press to alter an image along a sine wave.
Meta-P Press to simulate an oil painting.
Meta-C Press to simulate a charcoal drawing.
Alt-X Press to composite the image
with another.
Alt-A Press to annotate the image with text.
Alt-D Press to draw a line on the image.
Alt-P Press to edit an image pixel color.
Alt-M Press to edit the image matte information.
Alt-X Press to composite the image with another.
Alt-A Press to add a border to the image.
Alt-F Press to add a ornamental frame to the image.
Alt-Shft-! Press to add an image comment.
Ctl-A Press to apply image processing techniques to a
region of interest.
Shft-? Press to display information about the image.
Shft-+ Press to map the zoom image window.
Shft-P Press to preview an image enhancement, effect,
or f/x.
F1 Press to display helpful information about
the "display" utility.
Find Press to browse documentation about
GraphicsMagick.
1-9 Press to change the level of magnification.
Use the arrow keys to move the image one pixel up, down, left, or right within the magnify window. Be sure to first map the magnify window by pressing button 2.
Press ALT and one of the arrow keys to trim off one pixel from any side of the image.
Display options can appear on the command line or in your X resource file. Options on the command line supersede values specified in your X resource file. See X(1) for more information on X resources.
Most display options have a corresponding X resource. In addition, display uses the following X resources:
Offsets, if present, are handled in X(1) style. A negative x offset is measured from the right edge of the screen to the right edge of the icon, and a negative y offset is measured from the bottom edge of the screen to the bottom edge of the icon.
Offsets, if present, are handled in the same manner as in class Geometry.
To set the geometry of the Magnify or Pan or window, use the geometry resource. For example, to set the Pan window geometry to 256x256, use:
gm display.pan.geometry: 256x256
To select an image to display, choose Open of the File sub-menu from the Command widget. A file browser is displayed. To choose a particular image file, move the pointer to the filename and press any button. The filename is copied to the text window. Next, press Open or press the RETURN key. Alternatively, you can type the image file name directly into the text window. To descend directories, choose a directory name and press the button twice quickly. A scrollbar allows a large list of filenames to be moved through the viewing area if it exceeds the size of the list area.
You can trim the list of file names by using shell globbing characters. For example, type *.jpg to list only files that end with .jpg.
To select your image from the X server screen instead of from a file, Choose Grab of the Open widget.
To create a Visual Image Directory, choose Visual Directory of the File sub-menu from the Command widget . A file browser is displayed. To create a Visual Image Directory from all the images in the current directory, press Directory or press the RETURN key. Alternatively, you can select a set of image names by using shell globbing characters. For example, type *.jpg to include only files that end with .jpg. To descend directories, choose a directory name and press the button twice quickly. A scrollbar allows a large list of filenames to be moved through the viewing area if it exceeds the size of the list area.
After you select a set of files, they are turned into thumbnails and tiled onto a single image. Now move the pointer to a particular thumbnail and press button 3 and drag. Finally, select Open. The image represented by the thumbnail is displayed at its full size. Choose Next from the File sub-menu of the Command widget to return to the Visual Image Directory.
Note that cut information for image window is not retained for colormapped X server visuals (e.g. StaticColor, StaticColor, GRAYScale, PseudoColor). Correct cutting behavior may require a TrueColor or DirectColor visual or a Standard Colormap.
To begin, press choose Cut of the Edit sub-menu from the Command widget. Alternatively, press F3 in the image window.
A small window appears showing the location of the cursor in the image window. You are now in cut mode. In cut mode, the Command widget has these options:
Help
Dismiss
To define a cut region, press button 1 and drag. The cut region is defined by a highlighted rectangle that expands or contracts as it follows the pointer. Once you are satisfied with the cut region, release the button. You are now in rectify mode. In rectify mode, the Command widget has these options:
Cut
Help
Dismiss
You can make adjustments by moving the pointer to one of the cut rectangle corners, pressing a button, and dragging. Finally, press Cut to commit your copy region. To exit without cutting the image, press Dismiss.
To begin, press choose Copy of the Edit sub-menu from the Command widget. Alternatively, press F4 in the image window.
A small window appears showing the location of the cursor in the image window. You are now in copy mode. In copy mode, the Command widget has these options:
Help
Dismiss
To define a copy region, press button 1 and drag. The copy region is defined by a highlighted rectangle that expands or contracts as it follows the pointer. Once you are satisfied with the copy region, release the button. You are now in rectify mode. In rectify mode, the Command widget has these options:
Copy
Help
Dismiss
You can make adjustments by moving the pointer to one of the copy rectangle corners, pressing a button, and dragging. Finally, press Copy to commit your copy region. To exit without copying the image, press Dismiss.
To begin, press choose Paste of the Edit sub-menu from the Command widget. Alternatively, press F5 in the image window.
A small window appears showing the location of the cursor in the image window. You are now in Paste mode. To exit immediately, press Dismiss. In Paste mode, the Command widget has these options:
Operators
over
in
out
atop
xor
plus
minus
add
subtract
difference
multiply
bumpmap
replace
Help
Dismiss
Choose a composite operation from the Operators sub-menu of the Command widget. How each operator behaves is described below. image window is the image currently displayed on your X server and image is the image obtained with the File Browser widget.
The image compositor requires a matte, or alpha channel in the image for some operations. This extra channel usually defines a mask which represents a sort of a cookie-cutter for the image. This is the case when matte is 255 (full coverage) for pixels inside the shape, zero outside, and between zero and 255 on the boundary. If image does not have a matte channel, it is initialized with 0 for any pixel matching in color to pixel location (0,0), otherwise 255. See Matte Editing for a method of defining a matte channel.
Note that matte information for image window is not retained for colormapped X server visuals (e.g. StaticColor, StaticColor, GrayScale, PseudoColor). Correct compositing behavior may require a TrueColor or DirectColor visual or a Standard Colormap.
Choosing a composite operator is optional. The default operator is replace. However, you must choose a location to composite your image and press button 1. Press and hold the button before releasing and an outline of the image will appear to help you identify your location.
The actual colors of the pasted image is saved. However, the color that appears in image window may be different. For example, on a monochrome screen image window will appear black or white even though your pasted image may have many colors. If the image is saved to a file it is written with the correct colors. To assure the correct colors are saved in the final image, any PseudoClass image is promoted to DirectClass. To force a PseudoClass image to remain PseudoClass, use -colors.
To begin, press choose Crop of the Transform submenu from the Command widget. Alternatively, press C in the image window.
A small window appears showing the location of the cursor in the image window. You are now in crop mode. In crop mode, the Command widget has these options:
Help
Dismiss
To define a cropping region, press button 1 and drag. The cropping region is defined by a highlighted rectangle that expands or contracts as it follows the pointer. Once you are satisfied with the cropping region, release the button. You are now in rectify mode. In rectify mode, the Command widget has these options:
Crop
Help
Dismiss
You can make adjustments by moving the pointer to one of the cropping rectangle corners, pressing a button, and dragging. Finally, press Crop to commit your cropping region. To exit without cropping the image, press Dismiss.
An image is chopped interactively. There is no command line argument to chop an image. To begin, choose Chop of the Transform sub-menu from the Command widget. Alternatively, press [ in the Image window.
You are now in Chop mode. To exit immediately, press Dismiss. In Chop mode, the Command widget has these options:
Direction
horizontal
vertical
Help
Dismiss
If the you choose the horizontal direction (this is the default), the area of the image between the two horizontal endpoints of the chop line is removed. Otherwise, the area of the image between the two vertical endpoints of the chop line is removed.
Select a location within the image window to begin your chop, press and hold any button. Next, move the pointer to another location in the image. As you move a line will connect the initial location and the pointer. When you release the button, the area within the image to chop is determined by which direction you choose from the Command widget.
To cancel the image chopping, move the pointer back to the starting point of the line and release the button.
Press the / key to rotate the image 90 degrees or \ to rotate -90 degrees. To interactively choose the degree of rotation, choose Rotate... of the Transform submenu from the Command Widget. Alternatively, press * in the image window.
A small horizontal line is drawn next to the pointer. You are now in rotate mode. To exit immediately, press Dismiss. In rotate mode, the Command widget has these options:
Pixel Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Direction
horizontal
vertical
Help
Dismiss
Choose a background color from the Pixel Color sub-menu. Additional background colors can be specified with the color browser. You can change the menu colors by setting the X resources pen1 through pen9.
If you choose the color browser and press Grab, you can select the background color by moving the pointer to the desired color on the screen and press any button.
Choose a point in the image window and press this button and hold. Next, move the pointer to another location in the image. As you move a line connects the initial location and the pointer. When you release the button, the degree of image rotation is determined by the slope of the line you just drew. The slope is relative to the direction you choose from the Direction sub-menu of the Command widget.
To cancel the image rotation, move the pointer back to the starting point of the line and release the button.
An image is annotated interactively. There is no command line argument to annotate an image. To begin, choose Annotate of the Image Edit sub-menu from the Command widget. Alternatively, press a in the image window.
A small window appears showing the location of the cursor in the image window. You are now in annotate mode. To exit immediately, press Dismiss. In annotate mode, the Command widget has these options:
Font Name
fixed
variable
5x8
6x10
7x13bold
8x13bold
9x15bold
10x20
12x24
Browser...
Font Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Box Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Rotate Text
-90
-45
-30
0
30
45
90
180
Dialog...
Help
Dismiss
Choose a font name from the Font Name sub-menu. Additional font names can be specified with the font browser. You can change the menu names by setting the X resources font1 through font9.
Choose a font color from the Font Color sub-menu. Additional font colors can be specified with the color browser. You can change the menu colors by setting the X resources pen1 through pen9.
If you select the color browser and press Grab, you can choose the font color by moving the pointer to the desired color on the screen and press any button.
If you choose to rotate the text, choose Rotate Text from the menu and select an angle. Typically you will only want to rotate one line of text at a time. Depending on the angle you choose, subsequent lines may end up overwriting each other.
Choosing a font and its color is optional. The default font is fixed and the default color is black. However, you must choose a location to begin entering text and press a button. An underscore character will appear at the location of the pointer. The cursor changes to a pencil to indicate you are in text mode. To exit immediately, press Dismiss.
In text mode, any key presses will display the character at the location of the underscore and advance the underscore cursor. Enter your text and once completed press Apply to finish your image annotation. To correct errors press BACK SPACE. To delete an entire line of text, press DELETE. Any text that exceeds the boundaries of the image window is automatically continued onto the next line.
The actual color you request for the font is saved in the image. However, the color that appears in your Image window may be different. For example, on a monochrome screen the text will appear black or white even if you choose the color red as the font color. However, the image saved to a file with -write is written with red lettering. To assure the correct color text in the final image, any PseudoClass image is promoted to DirectClass (see miff(5)). To force a PseudoClass image to remain PseudoClass, use -colors.
An image composite is created interactively. There is no command line argument to composite an image. To begin, choose Composite of the Image Edit from the Command widget. Alternatively, press x in the Image window.
First a popup window is displayed requesting you to enter an image name. Press Composite, Grab or type a file name. Press Cancel if you choose not to create a composite image. When you choose Grab, move the pointer to the desired window and press any button.
If the Composite image does not have any matte information, you are informed and the file browser is displayed again. Enter the name of a mask image. The image is typically grayscale and the same size as the composite image. If the image is not grayscale, it is converted to grayscale and the resulting intensities are used as matte information.
A small window appears showing the location of the cursor in the image window. You are now in composite mode. To exit immediately, press Dismiss. In composite mode, the Command widget has these options:
Operators
over
in
out
atop
xor
plus
minus
add
subtract
difference
bumpmap
replace
Blend
Displace
Help
Dismiss
Choose a composite operation from the Operators sub-menu of the Command widget. How each operator behaves is described below. image window is the image currently displayed on your X server and image is the image obtained
The image compositor requires a matte, or alpha channel in the image for some operations. This extra channel usually defines a mask which represents a sort of a cookie-cutter for the image. This is the case when matte is 255 (full coverage) for pixels inside the shape, zero outside, and between zero and 255 on the boundary. If image does not have a matte channel, it is initialized with 0 for any pixel matching in color to pixel location (0,0), otherwise 255. See Matte Editing for a method of defining a matte channel.
If you choose blend, the composite operator becomes over. The image matte channel percent transparency is initialized to factor. The image window is initialized to (100-factor). Where factor is the value you specify in the Dialog widget.
Displace shifts the image pixels as defined by a displacement map. With this option, image is used as a displacement map. Black, within the displacement map, is a maximum positive displacement. White is a maximum negative displacement and middle gray is neutral. The displacement is scaled to determine the pixel shift. By default, the displacement applies in both the horizontal and vertical directions. However, if you specify mask, image is the horizontal X displacement and mask the vertical Y displacement.
Note that matte information for image window is not retained for colormapped X server visuals (e.g. StaticColor, StaticColor, GrayScale, PseudoColor). Correct compositing behavior may require a TrueColor or DirectColor visual or a Standard Colormap.
Choosing a composite operator is optional. The default operator is replace. However, you must choose a location to composite your image and press button 1. Press and hold the button before releasing and an outline of the image will appear to help you identify your location.
The actual colors of the composite image is saved. However, the color that appears in image window may be different. For example, on a monochrome screen Image window will appear black or white even though your composited image may have many colors. If the image is saved to a file it is written with the correct colors. To assure the correct colors are saved in the final image, any PseudoClass image is promoted to DirectClass (see miff). To force a PseudoClass image to remain PseudoClass, use -colors.
Changing the the color of a set of pixels is performed interactively. There is no command line argument to edit a pixel. To begin, choose Color from the Image Edit submenu of the Command widget. Alternatively, press c in the image window.
A small window appears showing the location of the cursor in the image window. You are now in color edit mode. To exit immediately, press Dismiss. In color edit mode, the Command widget has these options:
Method
point
replace
floodfill
reset
Pixel Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Border Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Fuzz
0
2
4
8
16
Dialog...
Undo
Help
Dismiss
Choose a color editing method from the Method sub-menu of the Command widget. The point method recolors any pixel selected with the pointer unless the button is released. The replace method recolors any pixel that matches the color of the pixel you select with a button press. Floodfill recolors any pixel that matches the color of the pixel you select with a button press and is a neighbor. Whereas filltoborder changes the matte value of any neighbor pixel that is not the border color. Finally reset changes the entire image to the designated color.
Next, choose a pixel color from the Pixel Color sub-menu. Additional pixel colors can be specified with the color browser. You can change the menu colors by setting the X resources pen1 through pen9.
Now press button 1 to select a pixel within the Image window to change its color. Additional pixels may be recolored as prescribed by the method you choose. additional pixels by increasing the Delta value.
If the Magnify widget is mapped, it can be helpful in positioning your pointer within the image (refer to button 2). Alternatively you can select a pixel to recolor from within the Magnify widget. Move the pointer to the Magnify widget and position the pixel with the cursor control keys. Finally, press a button to recolor the selected pixel (or pixels).
The actual color you request for the pixels is saved in the image. However, the color that appears in your Image window may be different. For example, on a monochrome screen the pixel will appear black or white even if you choose the color red as the pixel color. However, the image saved to a file with -write is written with red pixels. To assure the correct color text in the final image, any PseudoClass image is promoted to DirectClass To force a PseudoClass image to remain PseudoClass, use -colors.
Matte information within an image is useful for some operations such as image compositing. This extra channel usually defines a mask which represents a sort of a cookie-cutter for the image. This is the case when matte is 255 (full coverage) for pixels inside the shape, zero outside, and between zero and 255 on the boundary.
Setting the matte information in an image is done interactively. There is no command line argument to edit a pixel. To begin, and choose Matte of the Image Edit sub-menu from the Command widget.
Alternatively, press m in the image window.
A small window appears showing the location of the cursor in the image window. You are now in matte edit mode. To exit immediately, press Dismiss. In matte edit mode, the Command widget has these options:
Method
point
replace
floodfill
reset
Border Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Fuzz
0
2
4
8
16
Dialog...
Matte
Undo
Help
Dismiss
Choose a matte editing method from the Method sub-menu of the Command widget. The point method changes the matte value of the any pixel selected with the pointer until the button is released. The replace method changes the matte value of any pixel that matches the color of the pixel you select with a button press. Floodfill changes the matte value of any pixel that matches the color of the pixel you select with a button press and is a neighbor. Whereas filltoborder recolors any neighbor pixel that is not the border color. Finally reset changes the entire image to the designated matte value. Choose Matte Value and a dialog appears requesting a matte value. Enter a value between 0 and 255. This value is assigned as the matte value of the selected pixel or pixels. Now, press any button to select a pixel within the Image window to change its matte value. You can change the matte value of additional pixels by increasing the Delta value. The Delta value is first added then subtracted from the red, green, and blue of the target color. Any pixels within the range also have their matte value updated. If the Magnify widget is mapped, it can be helpful in positioning your pointer within the image (refer to button 2). Alternatively you can select a pixel to change the matte value from within the Magnify widget. Move the pointer to the Magnify widget and position the pixel with the cursor control keys. Finally, press a button to change the matte value of the selected pixel (or pixels). Matte information is only valid in a DirectClass image. Therefore, any PseudoClass image is promoted to DirectClass. Note that matte information for PseudoClass is not retained for colormapped X server visuals (e.g. StaticColor, StaticColor, GrayScale, PseudoColor) unless you immediately save your image to a file (refer to Write). Correct matte editing behavior may require a TrueColor or DirectColor visual or a Standard Colormap.
An image is drawn upon interactively. There is no command line argument to draw on an image. To begin, choose Draw of the Image Edit sub-menu from the Command widget. Alternatively, press d in the image window.
The cursor changes to a crosshair to indicate you are in draw mode. To exit immediately, press Dismiss. In draw mode, the Command widget has these options:
Primitive
point
line
rectangle
fill rectangle
circle
fill circle
ellipse
fill ellipse
polygon
fill polygon
Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Stipple
Brick
Diagonal
Scales
Vertical
Wavy
Translucent
Opaque
Open...
Width
1
2
4
8
16
Dialog...
Undo
Help
Dismiss
Choose a drawing primitive from the Primitive sub-menu.
Next, choose a color from the Color sub-menu. Additional colors can be specified with the color browser. You can change the menu colors by setting the X resources pen1 through pen9. The transparent color updates the image matte channel and is useful for image compositing.
If you choose the color browser and press Grab, you can select the primitive color by moving the pointer to the desired color on the screen and press any button. The transparent color updates the image matte channel and is useful for image compositing.
Choose a stipple, if appropriate, from the Stipple sub-menu. Additional stipples can be specified with the file browser. Stipples obtained from the file browser must be on disk in the X11 bitmap format.
Choose a line width, if appropriate, from the Width sub-menu. To choose a specific width select the Dialog widget.
Choose a point in the image window and press button 1 and hold. Next, move the pointer to another location in the image. As you move, a line connects the initial location and the pointer. When you release the button, the image is updated with the primitive you just drew. For polygons, the image is updated when you press and release the button without moving the pointer.
To cancel image drawing, move the pointer back to the starting point of the line and release the button.
To begin, press choose Region of Interest of the Pixel Transform sub-menu from the Command widget. Alternatively, press R in the image window.
A small window appears showing the location of the cursor in the image window. You are now in region of interest mode. In region of interest mode, the Command widget has these options:
Help
Dismiss
To define a region of interest, press button 1 and drag. The region of interest is defined by a highlighted rectangle that expands or contracts as it follows the pointer. Once you are satisfied with the region of interest, release the button. You are now in apply mode. In apply mode the Command widget has these options:
File
Save...
Print...
Edit
Undo
Redo
Transform
Flip
Flop
Rotate Right
Rotate Left
Enhance
Hue...
Saturation...
Brightness...
Gamma...
Spiff
Dull
Equalize
Normalize
Negate
GRAYscale
Quantize...
Effects
Despeckle
Emboss
Reduce Noise
Add Noise
Sharpen...
Blur...
Threshold...
Edge Detect...
Spread...
Shade...
Raise...
Segment...
F/X
Solarize...
Swirl...
Implode...
Wave...
Oil Paint
Charcoal Draw...
Miscellany
Image Info
Zoom Image
Show Preview...
Show Histogram
Show Matte
Help
Dismiss
You can make adjustments to the region of interest by moving the pointer to one of the rectangle corners, pressing a button, and dragging. Finally, choose an image processing technique from the Command widget. You can choose more than one image processing technique to apply to an area. Alternatively, you can move the region of interest before applying another image processing technique. To exit, press Dismiss.
When an image exceeds the width or height of the X server screen, display maps a small panning icon. The rectangle within the panning icon shows the area that is currently displayed in the the image window. To pan about the image, press any button and drag the pointer within the panning icon. The pan rectangle moves with the pointer and the image window is updated to reflect the location of the rectangle within the panning icon. When you have selected the area of the image you wish to view, release the button.
Use the arrow keys to pan the image one pixel up, down, left, or right within the image window.
The panning icon is withdrawn if the image becomes smaller than the dimensions of the X server screen.
Preferences affect the default behavior of display(1). The
preferences are either true or false and are stored in your home directory
as .displayrc:
display image centered on a backdrop"
This backdrop covers the entire workstation screen and is useful
for hiding other X window activity while viewing the image. The color of the
backdrop is specified as the background color. Refer to X Resources for
details.
confirm on program exit"
Ask for a confirmation before exiting the display(1)
program.
correct image for display gamma"
If the image has a known gamma, the gamma is corrected to match
that of the X server (see the X Resource displayGamma).
display warning messages"
Display any warning messages.
apply Floyd/Steinberg error diffusion to image"
The basic strategy of dithering is to trade intensity resolution
for spatial resolution by averaging the intensities of several neighboring
pixels. Images which suffer from severe contouring when reducing colors can
be improved with this preference.
use a shared colormap for colormapped X visuals"
This option only applies when the default X server visual is
PseudoColor or GRAYScale. Refer to -visual for more
details. By default, a shared colormap is allocated. The image shares colors
with other X clients. Some image colors could be approximated, therefore
your image may look very different than intended. Otherwise the image colors
appear exactly as they are defined. However, other clients may go
technicolor when the image colormap is installed.
display images as an X server pixmap"
Images are maintained as a XImage by default. Set this resource to
True to utilize a server Pixmap instead. This option is useful if your image
exceeds the dimensions of your server screen and you intend to pan the
image. Panning is much faster with Pixmaps than with a XImage. Pixmaps are
considered a precious resource, use them with discretion.
Identify describes the format and characteristics of one or more image files as internally supported by the software. It will also report if an image is incomplete or corrupt. The information displayed includes the scene number, the file name, the width and height of the image, whether the image is colormapped or not, the number of colors in the image, the number of bytes in the image, the format of the image (JPEG, PNM, etc.), and finally the number of seconds in both user time and elapsed time it took to read and process the image. If -verbose or +ping are provided as an option, the pixel read rate is also displayed. An example line output from identify follows:
images/aquarium.miff 640x480 PseudoClass 256c
308135b MIFF 0.000u 0:01
If -verbose is set, expect additional output including any image comment:
Image: images/aquarium.miff
class: PseudoClass
colors: 256
signature: eb5dca81dd93ae7e6ffae99a527eb5dca8...
matte: False
geometry: 640x480
depth: 8
bytes: 308135
format: MIFF
comments:
Imported from MTV raster image: aquarium.mtv
For some formats, additional format-specific information about the file will be written if the -debug coder or -debug all option is used.
Options are processed in command line order. Any option you specify on the command line remains in effect for the set of images immediately following, until the set is terminated by the appearance of any option or -noop.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
Import reads an image from any visible window on an X server and outputs it as an image file. You can capture a single window, the entire screen, or any rectangular portion of the screen. Use display for redisplay, printing, editing, formatting, archiving, image processing, etc. of the captured image.
The target window can be specified by id, name, or may be selected by clicking the mouse in the desired window. If you press a button and then drag, a rectangle will form which expands and contracts as the mouse moves. To save the portion of the screen defined by the rectangle, just release the button. The keyboard bell is rung once at the beginning of the screen capture and twice when it completes.
To select an X window or an area of the screen with the mouse and save it in the MIFF image format to a file entitled window.miff, use:
gm import window.miff
To select an X window or an area of the screen with the mouse and save it in the Encapsulated PostScript format to include in another document, use:
gm import figure.eps
To capture the entire X server screen in the JPEG image format in a file entitled root.jpeg, without using the mouse, use:
gm import -window root root.jpeg
To capture the 512x256 area at the upper right corner of the X server screen in the PNG image format in a well-compressed file entitled corner.png, without using the mouse, use:
gm import -window root -crop 512x256-0+0 -quality 90
corner.png
Options are processed in command line order. Any option you specify on the command line remains in effect until it is explicitly changed by specifying the option again with a different effect.
Import options can appear on the command line or in your X resources file. See X(1). Options on the command line supersede values specified in your X resources file.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
Mogrify transforms an image or a sequence of images. These transforms include image scaling, image rotation, color reduction, and others. Each transmogrified image overwrites the corresponding original image, unless an option such as -format causes the output filename to be different from the input filename.
The graphics formats supported by mogrify are listed in GraphicsMagick(1).
To convert all the TIFF files in a particular directory to JPEG, use:
gm mogrify -format jpeg *.tiff
To convert a directory full of JPEG images to thumbnails, use:
gm mogrify -size 120x120 *.jpg -resize 120x120 +profile "*"
In this example, '-size 120x120' gives a hint to the JPEG decoder that the images are going to be downscaled to 120x120, allowing it to run faster by avoiding returning full-resolution images to GraphicsMagick for the subsequent resizing operation. The ´-resize 120x120' specifies the desired dimensions of the output images. It will be scaled so its largest dimension is 120 pixels. The ´+profile "*"' removes any ICM, EXIF, IPTC, or other profiles that might be present in the input and aren't needed in the thumbnails.
To scale an image of a cockatoo to exactly 640 pixels in width and 480 pixels in height, use:
gm mogrify -resize 640x480! cockatoo.miff
Options are processed in command line order. Any option you specify on the command line remains in effect for the set of images that follows, until the set is terminated by the appearance of any option or -noop.
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
montage creates a composite image by combining several separate images. The images are tiled on the composite image with the name of the image optionally appearing just below the individual tile.
The composite image is constructed in the following manner. First, each image specified on the command line, except for the last, is scaled to fit the maximum tile size. The maximum tile size by default is 120x120. It can be modified with the -geometry command line argument or X resource. See Options for more information on command line arguments. See X(1) for more information on X resources. Note that the maximum tile size need not be a square.
Next the composite image is initialized with the color specified by the -background command line argument or X resource. The width and height of the composite image is determined by the title specified, the maximum tile size, the number of tiles per row, the tile border width and height, the image border width, and the label height. The number of tiles per row specifies how many images are to appear in each row of the composite image. The default is to have 5 tiles in each row and 4 tiles in each column of the composite. A specific value is specified with -tile. The tile border width and height, and the image border width defaults to the value of the X resource -borderwidth. It can be changed with the -borderwidth or -geometry command line argument or X resource. The label height is determined by the font you specify with the -font command line argument or X resource. If you do not specify a font, a font is chosen that allows the name of the image to fit the maximum width of a tiled area. The label colors is determined by the -background and -fill command line argument or X resource. Note, that if the background and pen colors are the same, labels will not appear.
Initially, the composite image title is placed at the top if one is specified (refer to -fill). Next, each image is set onto the composite image, surrounded by its border color, with its name centered just below it. The individual images are left-justified within the width of the tiled area. The order of the images is the same as they appear on the command line unless the images have a scene keyword. If a scene number is specified in each image, then the images are tiled onto the composite in the order of their scene number. Finally, the last argument on the command line is the name assigned to the composite image. By default, the image is written in the MIFF format and can be viewed or printed with display(1).
Note, that if the number of tiles exceeds the default number of 20 (5 per row, 4 per column), more than one composite image is created. To ensure a single image is produced, use -tile to increase the number of tiles to meet or exceed the number of input images.
Finally, to create one or more empty spaces in the sequence of tiles, use the "NULL:" image format.
Note, a composite MIFF image displayed to an X server with display behaves differently than other images. You can think of the composite as a visual image directory. Choose a particular tile of the composite and press a button to display it. See display(1) and miff(5)
To create a montage of a cockatoo, a parrot, and a hummingbird and write it to a file called birds, use:
gm montage cockatoo.miff parrot.miff hummingbird.miff
birds.miff
To tile several bird images so that they are at most 256 pixels in width and 192 pixels in height, surrounded by a red border, and separated by 10 pixels of background color, use:
gm montage -geometry 256x192+10+10 -bordercolor red
birds.* montage.miff
To create an unlabeled parrot image, 640 by 480 pixels, and surrounded by a border of black, use:
gm montage -geometry 640x480 -bordercolor black
-label "" parrot.miff bird.miff
To create an image of an eagle with a textured background, use:
gm montage -texture bumps.jpg eagle.jpg eagle.png
To join several GIF images together without any extraneous graphics (e.g. no label, no shadowing, no surrounding tile frame), use:
gm montage +frame +shadow +label -tile 5x1
-geometry 50x50+0+0 *.png joined.png
Any option you specify on the command line remains in effect for the group of images following it, until the group is terminated by the appearance of any option or -noop. For example, to make a montage of three images, the first with 32 colors, the second with an unlimited number of colors, and the third with only 16 colors, use:
gm montage -colors 32 cockatoo.1 -noop cockatoo.2
-colors 16 cockatoo.3 cockatoos.miff
For a more detailed description of each option, see Options, above.
For a more detailed description of each option, see Options, above.
Montage options can appear on the command line or in your X resource file. Options on the command line supersede values specified in your X resource file. See X(1) for more information on X resources.
All montage options have a corresponding X resource. In addition, montage uses the following X resources:
Specifies the preferred color to use for the composite image background. The default is #ccc.
Specifies the preferred color to use for the composite image border. The default is #ccc.
Specifies the width in pixels of the composite image border. The default is 2.
Specifies the name of the preferred font to use when displaying text within the composite image. The default is 9x15, fixed, or 5x8 determined by the composite image size.
Specify the color of an image frame. A 3D effect is achieved by using highlight and shadow colors derived from this color. The default value is #697B8F.
Specifies the preferred color to use for text within the composite image. The default is black.
This resource specifies the title to be placed at the top of the composite image. The default is not to place a title at the top of the composite image.
time executes an arbitrary gm utility command (e.g. convert) and reports the user and elapsed time. This provides way to measure command execution times similar to the Unix ´time' command but in a portable and consistent way.
To obtain time information for the execution of a command:
% gm time convert input.ppm -gaussian 0x2 output.ppm convert input.ppm -gaussian 0x2 output.ppm 22.60s user 0.00s system 2354% cpu 0.960 totalHere is the interpretation of the above output:
user - the total user time consumed.
system - the total system time consumed.
total - the total elapsed time consumed.
The time command reqires no options other than the gm command to execute.
version displays the software release version, build quantum (pixel sample) depth, web site URL, copyright notice, enabled features support, configuration parameters, and final build options used to build the software. The available information depends on how the software was configured and the host system.
To display the version information:
GraphicsMagick 1.3.37 2021-12-12 Q16 http://www.GraphicsMagick.org/
Copyright (C) 2002-2021 GraphicsMagick Group.
Additional copyrights and licenses apply to this software.
See http://www.GraphicsMagick.org/www/Copyright.html for details.
Feature Support:
Native Thread Safe yes
Large Files (> 32 bit) yes
Large Memory (> 32 bit) yes
BZIP yes
DPS no
FlashPix no
FreeType yes
Ghostscript (Library) no
JBIG yes
JPEG-2000 yes
JPEG yes
Little CMS yes
Loadable Modules no
Solaris mtmalloc no
Google perftools tcmalloc no
OpenMP yes (201511 "4.5")
PNG yes
TIFF yes
TRIO no
Solaris umem no
WebP yes
WMF yes
X11 yes
XML yes
ZLIB yes
Host type: x86_64-unknown-linux-gnu
Configured using the command:
./configure ...
Final Build Parameters:
CC = ...
CFLAGS = ...
CPPFLAGS = ...
CXX = ...
CXXFLAGS = ...
LDFLAGS = ...
LIBS = ...
The version command does not currently support any options.
| 2022/05/02 | GraphicsMagick |