Types of Filters

Here is a list of the filters that are always available in Imager. This list can be obtained by running the "filterlist.perl" script that comes with the module source.

  Filter          Arguments   Default value
  autolevels      lsat        0.1
                  usat        0.1
  autolevels_skew lsat        0.1
                  usat        0.1
                  skew        0
  bumpmap         bump lightx lighty
                  elevation   0
                  st          2
  bumpmap_complex bump
                  channel     0
                  tx          0
                  ty          0
                  Lx          0.2
                  Ly          0.4
                  Lz          -1 
                  cd          1.0 
                  cs          40.0
                  n           1.3
                  Ia          (0 0 0)
                  Il          (255 255 255)
                  Is          (255 255 255)
  contrast        intensity
  conv            coef
  fountain        xa ya xb yb
                  ftype        linear
                  repeat       none
                  combine      none
                  super_sample none
                  ssample_param 4
                  segments(see below)
  gaussian        stddev
  gaussian2       stddevX
                  stddevY
  gradgen         xo yo colors 
                  dist         0
  hardinvert
  hardinvertall
  mosaic          size         20
  noise           amount       3
                  subtype      0
  postlevels      levels       10
  radnoise        xo           100
                  yo           100
                  ascale       17.0
                  rscale       0.02
  turbnoise       xo           0.0
                  yo           0.0
                  scale        10.0
  unsharpmask     stddev       2.0
                  scale        1.0
  watermark       wmark
                  pixdiff      10
                  tx           0
                  ty           0

All parameters must have some value but if a parameter has a default value it may be omitted when calling the filter function.

Every one of these filters modifies the image in place.

If none of the filters here do what you need, the "transform()" in Imager::Engines or "transform2()" in Imager::Engines function may be useful.

A reference of the filters follows:

"autolevels"

Scales the luminance of the image so that the luminance will cover the possible range for the image. "lsat" and "usat" truncate the range by the specified fraction at the top and bottom of the range respectively.

  # increase contrast, losing little detail
  $img->filter(type=>"autolevels")
    or die $img->errstr;
    

The method used here is typically called Histogram Equalization <http://en.wikipedia.org/wiki/Histogram_equalization>.

"autolevels_skew"

Scales the value of each channel so that the values in the image will cover the whole possible range for the channel. "lsat" and "usat" truncate the range by the specified fraction at the top and bottom of the range respectively.

  # increase contrast per channel, losing little detail
  $img->filter(type=>"autolevels_skew")
    or die $img->errstr;
  # increase contrast, losing 20% of highlight at top and bottom range
  $img->filter(type=>"autolevels", lsat=>0.2, usat=>0.2)
    or die $img->errstr;
    

This filter was the original "autolevels" filter, but it's typically useless due to the significant color skew it can produce.

"bumpmap"

uses the channel "elevation" image "bump" as a bump map on your image, with the light at ("lightx", "lightty"), with a shadow length of "st".

  $img->filter(type=>"bumpmap", bump=>$bumpmap_img,
               lightx=>10, lighty=>10, st=>5)
    or die $img->errstr;
    

"bumpmap_complex"

uses the channel "channel" image "bump" as a bump map on your image. If "Lz < 0" the three L parameters are considered to be the direction of the light. If "Lz > 0" the L parameters are considered to be the light position. "Ia" is the ambient color, "Il" is the light color, "Is" is the color of specular highlights. "cd" is the diffuse coefficient and "cs" is the specular coefficient. "n" is the shininess of the surface.

  $img->filter(type=>"bumpmap_complex", bump=>$bumpmap_img)
    or die $img->errstr;
    

"contrast"

scales each channel by "intensity". Values of "intensity" < 1.0 will reduce the contrast.

  # higher contrast
  $img->filter(type=>"contrast", intensity=>1.3)
    or die $img->errstr;
  # lower contrast
  $img->filter(type=>"contrast", intensity=>0.8)
    or die $img->errstr;
    

"conv"

performs 2 1-dimensional convolutions on the image using the values from "coef". "coef" should be have an odd length and the sum of the coefficients must be non-zero.

  # sharper
  $img->filter(type=>"conv", coef=>[-0.5, 2, -0.5 ])
    or die $img->errstr;
  # blur
  $img->filter(type=>"conv", coef=>[ 1, 2, 1 ])
    or die $img->errstr;
  # error
  $img->filter(type=>"conv", coef=>[ -0.5, 1, -0.5 ])
    or die $img->errstr;
    

"fountain"

renders a fountain fill, similar to the gradient tool in most paint software. The default fill is a linear fill from opaque black to opaque white. The points "A(Cxa, ya)" and "B(xb, yb)" control the way the fill is performed, depending on the "ftype" parameter:

  # build the gradient the hard way - linear from black to white,
  # then back again
  my @simple =
   (
     [   0, 0.25, 0.5, 'black', 'white', 0, 0 ],
     [ 0.5. 0.75, 1.0, 'white', 'black', 0, 0 ],
   );
  # across
  my $linear = $img->copy;
  $linear->filter(type     => "fountain",
                  ftype    => 'linear',
                  repeat   => 'sawtooth',
                  segments => \@simple,
                  xa       => 0,
                  ya       => $linear->getheight / 2,
                  xb       => $linear->getwidth - 1,
                  yb       => $linear->getheight / 2)
    or die $linear->errstr;
  # around
  my $revolution = $img->copy;
  $revolution->filter(type     => "fountain",
                      ftype    => 'revolution',
                      segments => \@simple,
                      xa       => $revolution->getwidth / 2,
                      ya       => $revolution->getheight / 2,
                      xb       => $revolution->getwidth / 2,
                      yb       => 0)
    or die $revolution->errstr;
  # out from the middle
  my $radial = $img->copy;
  $radial->filter(type     => "fountain",
                  ftype    => 'radial',
                  segments => \@simple,
                  xa       => $im->getwidth / 2,
                  ya       => $im->getheight / 2,
                  xb       => $im->getwidth / 2,
                  yb       => 0)
    or die $radial->errstr;

"gaussian"

performs a Gaussian blur of the image, using "stddev" as the standard deviation of the curve used to combine pixels, larger values give bigger blurs. For a definition of Gaussian Blur, see:

  http://www.maths.abdn.ac.uk/~igc/tch/mx4002/notes/node99.html
    

Values of "stddev" around 0.5 provide a barely noticeable blur, values around 5 provide a very strong blur.

  # only slightly blurred
  $img->filter(type=>"gaussian", stddev=>0.5)
    or die $img->errstr;
  # more strongly blurred
  $img->filter(type=>"gaussian", stddev=>5)
    or die $img->errstr;
    

"gaussian2"

performs a Gaussian blur of the image, using "stddevX", "stddevY" as the standard deviation of the curve used to combine pixels on the X and Y axis, respectively. Larger values give bigger blurs. For a definition of Gaussian Blur, see:

  http://www.maths.abdn.ac.uk/~igc/tch/mx4002/notes/node99.html
    

Values of "stddevX" or "stddevY" around 0.5 provide a barely noticeable blur, values around 5 provide a very strong blur.

  # only slightly blurred
  $img->filter(type=>"gaussian2", stddevX=>0.5, stddevY=>0.5)
    or die $img->errstr;
  # blur an image in the Y axis
  $img->filter(type=>"gaussian", stddevX=>0, stddevY=>5 )
    or die $img->errstr;
    

"gradgen"

renders a gradient, with the given colors at the corresponding points (x,y) in "xo" and "yo". You can specify the way distance is measured for color blending by setting "dist" to 0 for Euclidean, 1 for Euclidean squared, and 2 for Manhattan distance.

  $img->filter(type="gradgen", 
               xo=>[ 10, 50, 10 ], 
               yo=>[ 10, 50, 50 ],
               colors=>[ qw(red blue green) ]);
    

"hardinvert"

inverts the image, black to white, white to black. All color channels are inverted, excluding the alpha channel if any.

  $img->filter(type=>"hardinvert")
    or die $img->errstr;
    

"hardinvertall"

inverts the image, black to white, white to black. All channels are inverted, including the alpha channel if any.

  $img->filter(type=>"hardinvertall")
    or die $img->errstr;
    

"mosaic"

produces averaged tiles of the given "size".

  $img->filter(type=>"mosaic", size=>5)
    or die $img->errstr;
    

"noise"

adds noise of the given "amount" to the image. If "subtype" is zero, the noise is even to each channel, otherwise noise is added to each channel independently.

  # monochrome noise
  $img->filter(type=>"noise", amount=>20, subtype=>0)
    or die $img->errstr;
  # color noise
  $img->filter(type=>"noise", amount=>20, subtype=>1)
    or die $img->errstr;
    

"radnoise"

renders radiant Perlin turbulent noise. The center of the noise is at ("xo", "yo"), "ascale" controls the angular scale of the noise , and "rscale" the radial scale, higher numbers give more detail.

  $img->filter(type=>"radnoise", xo=>50, yo=>50,
               ascale=>1, rscale=>0.02)
    or die $img->errstr;
    

"postlevels"

alters the image to have only "levels" distinct level in each channel.

  $img->filter(type=>"postlevels", levels=>10)
    or die $img->errstr;
    

"turbnoise"

renders Perlin turbulent noise. ("xo", "yo") controls the origin of the noise, and "scale" the scale of the noise, with lower numbers giving more detail.

  $img->filter(type=>"turbnoise", xo=>10, yo=>10, scale=>10)
    or die $img->errstr;
    

"unsharpmask"

performs an unsharp mask on the image. This increases the contrast of edges in the image.

This is the result of subtracting a Gaussian blurred version of the image from the original. "stddev" controls the "stddev" parameter of the Gaussian blur. Each output pixel is:

  in + scale * (in - blurred)
    

eg.

  $img->filter(type=>"unsharpmask", stddev=>1, scale=>0.5)
    or die $img->errstr;
    

"unsharpmark" has the following parameters:

"watermark"

applies "wmark" as a watermark on the image with strength "pixdiff", with an origin at ("tx", "ty")

  $img->filter(type=>"watermark", tx=>10, ty=>50, 
               wmark=>$wmark_image, pixdiff=>50)
    or die $img->errstr;
    

A demonstration of most of the filters can be found at:

  http://www.develop-help.com/imager/filters.html

External Filters

As of Imager 0.48 you can create perl or XS based filters and hook them into Imager's filter() method:

register_filter()

Registers a filter so it is visible via Imager's filter() method.

  Imager->register_filter(type => 'your_filter',
                          defaults => { parm1 => 'default1' },
                          callseq => [ qw/image parm1/ ],
                          callsub => \&your_filter);
  $img->filter(type=>'your_filter', parm1 => 'something');
    

The following parameters are needed:

Plug-ins

The plug in interface is deprecated. Please use the Imager API, see Imager::API and "External Filters" for details

It is possible to add filters to the module without recompiling Imager itself. This is done by using DSOs (Dynamic shared object) available on most systems. This way you can maintain your own filters and not have to have it added to Imager, or worse patch every new version of Imager. Modules can be loaded AND UNLOADED at run time. This means that you can have a server/daemon thingy that can do something like:

  load_plugin("dynfilt/dyntest.so")
    or die "unable to load plugin\n";
  $img->filter(type=>'lin_stretch', a=>35, b=>200);
  unload_plugin("dynfilt/dyntest.so")
    or die "unable to load plugin\n";

Someone decides that the filter is not working as it should - dyntest.c can be modified and recompiled, and then reloaded:

  load_plugin("dynfilt/dyntest.so")
    or die "unable to load plugin\n";
  $img->filter(%hsh);

Note: This has been tested successfully on the following systems: Linux, Solaris, HPUX, OpenBSD, FreeBSD, TRU64/OSF1, AIX, Win32, OS X.

load_plugin()

This is a function, not a method, exported by default. You should import this function explicitly for future compatibility if you need it.

Accepts a single parameter, the name of a shared library file to load.

Returns true on success. Check Imager->errstr on failure.

unload_plugin()

This is a function, not a method, which is exported by default. You should import this function explicitly for future compatibility if you need it.

Accepts a single parameter, the name of a shared library to unload. This library must have been previously loaded by load_plugin().

Returns true on success. Check Imager->errstr on failure.

A few example plug-ins are included and built (but not installed):

• plugins/dyntest.c - provides the "null" (no action) filter, and "lin_stretch" filters. "lin_stretch" stretches sample values between "a" and "b" out to the full sample range.
• plugins/dt2.c - provides the "html_art" filter that writes the image to the HTML fragment file supplied in "fname" as a HTML table.
• plugins/flines.c - provides the "flines" filter that dims alternate lines to emulate an old CRT display. Imager::Filter::Flines provides the same functionality.
• plugins/mandelbrot.c - provides the "mandelbrot" filter that renders the Mandelbrot set within the given range of x [-2, 0.5) and y [-1.25, 1,25). Imager::Filter::Mandelbrot provides a more flexible Mandelbrot set renderer.

Image Difference

difference()

You can create a new image that is the difference between 2 other images.

  my $diff = $img->difference(other=>$other_img);
    

For each pixel in $img that is different to the pixel in $other_img, the pixel from $other_img is given, otherwise the pixel is transparent black.

This can be used for debugging image differences ("Where are they different?"), and for optimizing animated GIFs.

Note that $img and $other_img must have the same number of channels. The width and height of $diff will be the minimum of each of the width and height of $img and $other_img.

Parameters:

rgb_difference()

You can create a new image that is the difference between 2 other images.

  my $diff = $img->rgb_difference(other=>$other_img);
    

For each pixel in $img that is different to the pixel in $other_img, the arithmetic difference for the value of the pixel in $img from $other_img per color is given. Transparency is ignored.

This can be used for measuring image differences ("How much are they different?").

Note that $img and $other_img must have the same number of channels. The width and height of $diff will be the minimum of each of the width and height of $img and $other_img.

Parameters: