文档介绍：第 28 卷第 9 .9
2011 年 9 月 Vol .2011No
Application Research puters Sep
基于压缩感知理论的分类量化图像编码方法倡
张伟, 曾凡仔, 曾庆光
(湖南大学信息科学与工程学院, 长沙 410082)
摘要: 针对矢量量化压缩速度慢、图像复原效果不理想等问题,根据图像小波分解后高频子带稀疏的特点,提
出了一种基于压缩感知( , ) 理论的分类量化图像编码算法。仿真结果表明,与矢量量
compressed sensing CS LBG
化编码算法相比,重构图像质量得到极大提升,在相似压缩比下,该算法取得了较好的效果, 平均有 1 ~3
PSNR
的明显提高;在相似信噪比( )下,该算法在图像压缩方面也有很大改进。
dB PSNR
关键词: 压缩感知; 分类量化编码; 随机观测; 优化问题
中图分类号: 文献标志码: 文章编号: 1001唱3695(2011)09唱3595唱03
TN A
: /. .
doi j issn
Method of categories quantization coding for image based pressed sensing
, 唱, 唱
ZHANG Wei ZENG Fan zi ZENG Qing guang
( School munication, Hunan University, Changsha 410082, China)
&
Abstract: 唱
, For the problem in vector quantization that pression ratio is slow and image restoration, result is not satisfac
tory and according to the sparsity properties of the high frequency wavelet transform. coefficients the paper proposed a new
method of categories, quantization coding for image based pressed pared with the LBG vector quantization唱
coding. algorithm simulation results demonstrate, that the proposed algorithm improves the quality of13 the recovered. image signifi,
cantly For the pression ratio the PSNR of the proposed. algorithm was improved about dB For the similar PSNR
theKey algorithm words: in pression; has also improved significant