blender(3x) | AfterStep X11 window manager | blender(3x) |
blender - functionality for blending of image data using diofferent algorithms libAfterImage/blender.h
Defines implemented methods for ASScanline combining, that could
be passed to merge_layers() via ASImageLayer structure. Also includes
functions for colorspace conversion RGB<->HSV and RGB<->HLS.
Functions :
Colorspace conversion :
rgb2value(), rgb2saturation(), rgb2hue(), rgb2luminance(),
rgb2hsv(), rgb2hls(), hsv2rgb(), hls2rgb().
merge_scanline methods :
alphablend_scanlines(), allanon_scanlines(),
tint_scanlines(), add_scanlines(), sub_scanlines(),
diff_scanlines(), darken_scanlines(), lighten_scanlines(),
screen_scanlines(), overlay_scanlines(), hue_scanlines(),
saturate_scanlines(), value_scanlines(),
colorize_scanlines(), dissipate_scanlines().
useful merging function name to function translator :
blend_scanlines_name2func()
Other libAfterImage modules :
ascmap.h asfont.h asimage.h asvisual.h blender.h export.h
import.h transform.h ximage.h
Sasha Vasko <sasha at aftercode dot net>
libAfterImage/colorspace
RGB colorspace: each color is represented as a combination of red, green and blue values. Each value can be in 2 formats : 8 bit and 24.8 bit. 24.8 bit makes for 32bit value with lower 8 bits used as a fraction for better calculation precision.
HSV colorspace: each color is represented as a combination of hue, saturation and value. Hue is generally colorizing component where value represents brightness.
HLS colorspace: each color is represented as a combination of hue, luminance and saturation. It is analogous to HSV with value substituted by luminance, except that luminance could be both negative and positive.
alpha channel could be added to any of the above colorspaces.
alpha channel is generally used to define transparentness of the color.
libAfterImage is using ARGB colorspace as a base colorspace, and represents
most colors as ARGB32 values or ASScanline scanlines of pixels.
libAfterImage/rgb2value()
rgb2saturation()
rgb2hue()
rgb2luminance()
CARD32 rgb2value( CARD32 red, CARD32 green, CARD32 blue ); CARD32
rgb2saturation( CARD32 red, CARD32 green, CARD32 blue ); CARD32 rgb2hue(
CARD32 red, CARD32 green, CARD32 blue ); CARD32 rgb2luminance (CARD32 red,
CARD32 green, CARD32 blue );
32 bit value, 16 lower bits of which represent value, saturation,
hue, or luminance respectively.
This functions translate RGB color into respective coordinates of
HSV and HLS colorspaces. Returned hue values are in 16bit format. To
translate it to and from conventional 0-360 degree range, please use :
degrees2hue16() - converts conventional hue in 0-360 range into hue16
hue162degree() - converts 16bit hue value into conventional degrees.
libAfterImage/rgb2hsv()
rgb2hls()
CARD32 rgb2hsv( CARD32 red, CARD32 green, CARD32 blue,
CARD32 *saturation, CARD32 *value ); CARD32 rgb2hls( CARD32 red, CARD32
green, CARD32 blue,
CARD32 *luminance, CARD32 *saturation );
32 bit value, 16 lower bits of which represent hue. 32bit value
pointed to by luminance, value and saturation will be set respectively to
color luminance, value and saturation.
This functions translate RGB color into full set of HSV and HLS
coordinates at once. These functions work faster then separate translation
into each channel.
libAfterImage/hsv2rgb()
hls2rgb()
void hsv2rgb( CARD32 hue, CARD32 saturation, CARD32 value,
CARD32 *red, CARD32 *green, CARD32 *blue); void hls2rgb( CARD32 hue, CARD32
luminance, CARD32 saturation,
CARD32 *red, CARD32 *green, CARD32 *blue);
32bit value pointed to by red, green and blue will be set
respectively to RGB color channels.
This functions performs reverse translation from HSV and HSL to
RGB color
libAfterImage/merge_scanline
- combines top and bottom RGB components based on alpha channel value: bottom = bottom*(255-top_alpha)+top*top_alpha; allanon_scanlines() - averages each pixel between two scanlines. This method has been first implemented by Ethan Fisher aka allanon as mode 130: bottom = (bottom+top)/2; tint_scanlines() - tints bottom scanline with top scanline( with saturation to prevent overflow) : bottom = (bottom*(top/2))/32768; add_scanlines() - adds top scanline to bottom scanline with saturation to prevent overflow: bottom = bottom+top; sub_scanlines() - substrates top scanline from bottom scanline with saturation to prevent overflow: bottom = bottom-top; diff_scanlines() - for each pixel calculates absolute difference between bottom and top color value : bottom = (bottom>top)?bottom-top:top-bottom; darken_scanlines() - substitutes each pixel with minimum color value of top and bottom : bottom = (bottom>top)?top:bottom; lighten_scanlines() - substitutes each pixel with maximum color value of top and bottom : bottom = (bottom>top)?bottom:top; screen_scanlines() - some weird merging algorithm taken from GIMP; overlay_scanlines() - some weird merging algorithm taken from GIMP; hue_scanlines() - substitute hue of bottom scanline with hue of top scanline; saturate_scanlines() - substitute saturation of bottom scanline with the saturation of top scanline; value_scanlines() - substitute value of bottom scanline with the value of top scanline; colorize_scanlines() - combine luminance of bottom scanline with hue and saturation of top scanline; dissipate_scanlines()- randomly alpha-blend bottom and top scanlines, using alpha value of top scanline as a threshold for random values.
void alphablend_scanlines( ASScanline *bottom, ASScanline *top,
int ); void allanon_scanlines ( ASScanline *bottom, ASScanline *top, int );
void tint_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
add_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
sub_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
diff_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
darken_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
lighten_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
screen_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
overlay_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
hue_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
saturate_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
value_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
colorize_scanlines ( ASScanline *bottom, ASScanline *top, int ); void
dissipate_scanlines ( ASScanline *bottom, ASScanline *top, int );
This functions accept 2 scanlines as an arguments stored in
ASScanline structures with data in 24.8 format. Merging operation is
performed on these scanlines and result is stored in bottom ASScanline. The
following are merging methods used in each function :
libAfterImage/blend_scanlines_name2func()
list_scanline_merging()
merge_scanlines_func blend_scanlines_name2func( const char *name
); void list_scanline_merging(FILE* stream, const char *format);
returns pointer to the scanline merging function on succes. NULL
on failure.
blend_scanlines_name2func() will strip leading whitespaces off of
the supplied name, and then will attempt to match it against the list of
names of merging functions. It will then return pointer to the function with
matching name. list_scanline_merging() simply prints out description of
implemented blending/merging methods onto the supplied stream, in supplied
format. Format must include 2 string specs, like so : "%s - %s"
where first one will be substituted to short method name, and second -
description
AfterStep v.2.2.12 | 3rd Berkeley Distribution |