ImRdoptCompress is a tcl interface to ImRdopt for
compressing as closely to a target size (or target bpp)
as possible. The script calls ImRdopt, so make sure
it is on your path.
Usage:
Pass the entire command-line as you would for ImRdopt.
ImRdoptCompress can accept some additional flags, which
are explained below:
--size
--bpp
Specify what kind of target you want: size or bpp.
Default is bpp, and the default target is 1.0 (see --target)
--target T
Specify that you want target T (which is either a size in
bytes or a bpp value, determined by --size or --bpp).
Default is 1.0
--attempts N
How many attempts to make to acheive the target.
Default is 5.
The program also terminates if it starts cycling.
--tolerance tol
--postolerance ptol
--negtolerance ntol
--bestabserr
How much error you are willing to accept in achieving the
target. Default is 0. You can set this to a higher value,
and increase "attempts". --tolerance sets both positive
and negative tolerance to tol. If the target cannot be achieved
within the prescribed tolerance, by default, the best result
*under* the target will be used. This behavior can be changed
by specifying --bestabserr, in which case the the case that
gave the best absolute error will be used.
--outfile fname
The compressed file will be "fname".
Default is cfile.jpg
--sequential
--progressive
Determines sequential or progressive JPEG. Default is sequential.
Example
---------If you have an image foo.ppm that you want to compress to a
sequntial JPEG file foo.jpg with size nearly 10000 bytes, you
might do:
ImRdoptCompress -coarse 4 -thresh 20 foo.ppm --size --target 10000 \
--attempts 10 --tolerance 50 --outfile foo.jpg --sequential