项目作者: xdsopl

项目描述 :
Playing with lossy image compression based on the singular value decomposition
高级语言: C
项目地址: git://github.com/xdsopl/svd.git
创建时间: 2021-05-25T20:55:36Z
项目社区:https://github.com/xdsopl/svd

开源协议:BSD Zero Clause License

下载


Playing with lossy image compression based on the singular value decomposition

Quick start:

Encode smpte.ppm PNM picture file to encoded.svd:

  1. ./encode smpte.ppm encoded.svd

Decode encoded.svd file to decoded.ppm picture file:

  1. ./decode encoded.svd decoded.ppm

Watch decoded.ppm picture file in feh:

  1. feh decoded.ppm

Adjusting quantization

Use quantization values of seven for luminance (Y’), six and seven for chrominance (Cb and Cr) instead of the default 12 10 10 values:

  1. ./encode smpte.ppm encoded.svd 7 6 5

Limited storage capacity

Use up to 65536 bits of space instead of the default 0 (no limit) and discard quality bits, if necessary, to stay below 65536 bits:

  1. ./encode smpte.ppm encoded.svd 12 10 10 65536

Reading

Singular value decomposition