一种图像自适应小波压缩算法  

A Novel Image Compression Algorithm Based on the Wavelet Transform

作　　者：陈旻[1] 王开云[2] 薛洁[3] 罗迪[1] 

机构地区：[1]云南大学信息学院,云南昆明650061 [2]昆明学院学报编辑部,云南昆明650214 [3]云南警官学院信息网络安全学院,云南昆明650223

出　　处：《昆明学院学报》2013年第6期96-99,共4页Journal of Kunming University

基　　金：云南省自然科学基金青年基金资助项目(2013FD042)

摘　　要：给出一种自适应图像小波压缩算法,首先将图像进行小波变换.然后将得到的小波系数进行均匀量化,并将量化后的系数分解为3个部分,对不同部分建立不同的Context模型,然后分别进行编码.为缓解模型稀释效应,使算法自适应的获得最优Context量化级数,使用了基于最短自适应码长增量的Context量化方法.结果表明,本算法不仅能自适应的获得最优量化级数,使编码码长最短,同时能够保证较好的图像信噪比,与其它小波压缩算法比较,进一步提高了压缩效率.A novel image compression algorithm based on the wavelet transform is discussed. The wavelet transform operation is de- ployed firstly for the image waiting for being coded and the uniform quantization is used to reduce the spatial cost for the coding model subsequently. Then the coefficients of the wavelet transfortning are decomposed into three parts. The Context models are established for coding these partitioned coefficients separately to find the optimal Context quantization levels. The context quantization based on the minimum increment of the adaptive code length is proposed to tackle the ＂model dilution＂. The experiment results indicate that the proposed algorithm not only gets the optimal quantization levels adaptively with the minimum code length but also keeps better image signal-to-interference ratio. Compared with other algorithms, the better compression efficiency is improved.

关 键 词：图像压缩 小波变换 Context熵编码 自适应码长增量 

分 类 号：TP919.1[自动化与计算机技术]