pixshrink [-u] [-w _in_width_] [-n _in_height_] [-s _squaresize_] [-f _shrink_factor_] [_infile.pix_] > _outfile.pix_PIXSHRINK(1)
NAME
pixshrink - shrink a color pix file by an integer factor
On this page
SYNOPSIS
DESCRIPTION
pixshrink reads a color pix(5) format image of given dimensions and produces a smaller, scaled-down version. The output is reduced by an integer shrink_factor in both dimensions, making pixshrink the logical opposite of pixinterp2x(1).
Input is read from infile.pix if given, otherwise from standard input; the shrunk image is written to standard output.
The input width and height (in pixels) are specified with -w and -n. Both default to 512 when not given. The entire image is read into memory, then scaled. By default pixshrink uses boxcar averaging, computing each output pixel as the average of a shrink_factor by shrink_factor block of input pixels; the -u option selects faster undersampling (point sampling) instead.
OPTIONS
- -w in_width
-
Set the input image width in pixels (default 512).
- -n in_height
-
Set the input image height in scan lines (default 512).
- -s squaresize
-
Set both the input width and height to squaresize pixels (shorthand for a square image).
- -f shrink_factor
-
Set the integer scaling factor (default 2). Each output dimension becomes 1/shrink_factor of the input. Must be greater than 1.
- -u
-
Use image undersampling (point sampling) rather than the default boxcar averaging.
- -h, -?
-
Print a usage message and exit.
EXAMPLES
Example 1. Halve a 1024x1024 image
pixshrink -s 1024 image.pix > thumb.pixReads a 1024x1024 pix(5) image and writes a 512x512 version using the default shrink factor of 2 and boxcar averaging.
Example 2. Generate a quarter-size thumbnail and view it
pixshrink -w 1280 -n 720 -f 4 render.pix | pix-fb -w 320 -n 180Shrinks a 1280x720 rendering to 320x180 (factor of 4) and pipes the result straight to pix-fb(1) for a quick preview on the framebuffer, without writing an intermediate file.
BUGS
If the input dimensions are not evenly divisible by the shrink_factor, incorrect or garbled output is produced.
SEE ALSO
brlcad(1), pixinterp2x(1), pixscale(1), pixcrop(1), pixrect(1), pix-fb(1), pix(5)
AUTHOR
BRL-CAD Team
COPYRIGHT
This software is Copyright (c) 1990-2026 by the United States Government as represented by U.S. Army Research Laboratory.
BUG REPORTS
Reports of bugs or problems should be submitted via electronic mail to devs@brlcad.org