Hacker News new | ask | show | jobs
by abainbridge 3114 days ago
Yeah, you can use principle component analysis to figure out the transform to make the colour space "optimal" for compression of a specific image. I tried that once:

https://bainbridgecode.wordpress.com/2012/04/24/beating-png-...

and

https://bainbridgecode.wordpress.com/2012/05/07/beating-png-...

There are plots at the end of part 3 showing the image channels separately for YCrCb and my custom PCA derived space. They clearly show that there's a lot less duplication of data between channels with the latter. And that in turn is obviously good for compression.

1 comments

Finding the first primary component also appears to be a pretty good way greyscale an image.