Skip to content

SunYanCN/Image-Tucker-Compress

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Image-Tucker-Compress

分别使用tensorly和tensorcomlib测试了张量图像压缩

安装Python库

numpy,pillow,matplotlib,tensorly,skimage,tensorcomlib

Tensorly测试结果

运行tensorly文件夹下的image_tucker.py文件,设置tucker_rank分别为[50,50,3]、[100,100,3]、[200,200,3]和[300,300,3]。如果报Memory错误,修改tensorly源代码,full_matrx = False

运行结果如下:

rank = [50,50,3]

Image Compression Ratio:0.925347646077474
Image Compare PSNR:25.946412228266816
Decomposition Time:0.7083556652069092

rank = [100,100,3]

Image Compression Ratio:0.8316332499186198
Image Compare PSNR:28.686432064795255
Decomposition Time:0.8138234615325928

rank = [200,200,3]

Image Compression Ratio:0.5869839986165364
Image Compare PSNR:30.804968069749638
Decomposition Time:1.4251620769500732

rank = [300,300,3]

Image Compression Ratio:0.2660408020019531
Image Compare PSNR:38.38228135960186
Decomposition Time:1.942857027053833

Tensorcomlib

Tensorcomlib是我自己练习写的一个库,测试效果和tensorly相似,输出信息可能更详细一点。

比较

联系方式

sunyanqinyin@foxmail.com

参考资料

  1. Tensorly文档
  2. 《基于张量Tucker分解的彩色图像压缩 》,王东方

Releases

No releases published

Packages

No packages published

Languages