2000.10.20 18:50 "YCbCr and Testing", by Frank Warmerdam


I believe I have fixed all the problems related to YCbCr handling and images (and strip sizes) that are not multiples of the sampling size. Anyone into YCbCr support is encouraged to try out the code in CVS. This included changes to the packbits code, and rgb2ycbcr. Files were being generated with odd amounts of data in the strips... more than there should have been.

Through this process, I have been wishing we had decent automated test streams. If there is anyone out there who would like to contribute to libtiff, we could really use some contributed test streams. I was thinking something along the lines of:

Anyone interested in taking this on should contact me by email.

Best regards,

I set the clouds in motion - turn up   | Frank Warmerdam, warmerda@home.com
light and sound - activate the windows | http://members.home.com/warmerda
and watch the world go round - Rush    | Geospatial Programmer for Rent