This is documentation for Mathematica 7, which was
based on an earlier version of the Wolfram Language.

# ImageCompose

 ImageCompose[image, overlay]gives the result of overlaying obj onto image. ImageCompose[image, {overlay, }]gives the result of alpha-blending overlay into image using blending fraction . ImageCompose[image, overlay, pos]places the center of overlay at position pos in image. ImageCompose[image, overlay, pos, opos]places the point opos in overlay at position pos in image. ImageCompose[image, overlay, pos, opos, {fi, fo, mode}]uses the compositing fractions fk and the specified compositing mode.
• ImageCompose[image, ...] always gives an image with the same pixel size as image.
• ImageCompose[image, obj] does alpha-blending if obj has an alpha channel.
• ImageCompose[image, {obj, }] effectively sets the alpha value for obj to , or multiplies an existing alpha channel by .
• ImageCompose[image, obj] overlays obj pixel-by-pixel onto image, without changing the size of obj.
• The object obj can be either an explicit image object, or a graphics object that is automatically rasterized to have the same image size as image.
• The position pos in image, or the position opos in obj can be specified in the following ways:
 {x,y} absolute pixel offset from the bottom-left corner Scaled[{sx,sy}] scaled position from 0 to 1 across the object {posx,posy} named x and y positions
• If only one of posx and posy is given, the other is assumed to be Center.
• The pixel offsets {x, y} can be negative, indicating a point outside of image or obj.
• ImageCompose[image, ...] yields an image with an alpha channel only if image has an alpha channel.
• ImageCompose[{image, i}, ...] effectively adds an alpha channel with value i to image.
• ImageCompose[..., {fi, fo, mode}] combines a fraction fi of every pixel value in image with a fraction fo of every pixel value in obj, in all cases taking account of alpha channels.
• With value 1 for mode, obj is treated as on top of image. With value -1 it is treated as underneath. And with value 0, image and obj are treated symmetrically.
• The following choices of {fi, fo, mode} correspond to standard Duff-Porter compositing operators:
 {1,1,1} object "over" image (default) {0,1,1} object "atop" image {1,0,1} object {0,0,1} object "in" image {1,1,0} "xor" operation {0,1,0} image "out" object {1,0,0} object "out" image {0,0,0} "clear" image {1,1,-1} image "over" object {0,1,-1} image {1,0,-1} image "atop" object {0,0,-1} image "in" object
• The "over" operators given by {1, 1, 1} and {1, 1, -1} effectively perform alpha blending.
• The "in" operators given by {0, 0, 1} and {0, 0, -1} mask one image with the alpha channel of the other image.
• The "out" operators {0, 1, 0} and {1, 0, 0} mask one image with the complement of the alpha channel of the other image.
• The "atop" operators given by {0, 1, 1} and {1, 0, -1} effectively combine the alpha channels of the two images before performing alpha blending.
Overlay an image with transparent text:
Specify an explicit position for the overlay:
Specify an alpha blending parameter:
Overlay two partially transparent images:
Overlay an image with transparent text:
 Out[1]=

Specify an explicit position for the overlay:
 Out[1]=

Specify an alpha blending parameter:
 Out[1]=

Overlay two partially transparent images:
 Out[1]=
 Applications   (1)
Create a motion blur effect:
Explore the full range of alpha-compositing modes and parameters:
New in 7