基于图像压缩算法的计算复杂度的讨论  

Computation Complexity Based on Image Compression Algorithm

作　　者：朱里[1] 张婷[1] 王文涛[1] 

机构地区：[1]华中科技大学图像识别与人工智能研究所生物与医学图像研究室,武汉430074

出　　处：《光学与光电技术》2007年第4期24-27,共4页Optics & Optoelectronic Technology

基　　金：国家自然科学基金(60372066)资助项目

摘　　要：多级树集合分裂算法是一种简单有效的嵌入式零树编码算法,但它需要多次重复运算,复杂程度高,降低了编码效率。针对该算法的不足,提出了一种改进的多级树集合分裂算法,在相同的小波变换条件下,通过增加零树深度,减少了表示位置信息的系数,提高了表示有效系数的信息,该算法改变了原来的扫描顺序,便于并行处理。实验表明在压缩性能相同的情况下,提高了编码效率。SPIHT algorithm is an easy and effective embedded zero tree coding algorithm. But it needs many repeated calculation and is more complicated, which decreases the coding efficiency. An improved SPIHT algorithm is proposed according to the disadvantage of the former. Effective coefficient information is improved by increasing the zero tree depth and decreasing the position information coefficient under the same conditon of wavelet transformation. The algorithm changes the original scanning sequence in order for parallel processing. Experiments show that it improves the coding efficiency under the same condition of compression performance.

关 键 词：零树深度 有效系数 控制比特 

分 类 号：TP301.5[自动化与计算机技术—计算机系统结构]