首页 | 本学科首页   官方微博 | 高级检索  
     检索      

WHT图象压缩及增强
引用本文:蔡德钧,唐震宇.WHT图象压缩及增强[J].华中科技大学学报(自然科学版),1991(Z1).
作者姓名:蔡德钧  唐震宇
作者单位:华中理工大学电子与信息工程系 (蔡德钧),华中理工大学电子与信息工程系(唐震宇)
基金项目:国家自然科学基金资助项目
摘    要:本文提出一种Walsh-Hadamard变换(以下简写为WHT)图象压缩方法.该方法将正交变换后的系数矩阵按活性指数分为三类,然后进行自适应门限压缩、比特分配与量化.文中分析了设计原理并给出了实验结果.为了改善低码率下编码图象的质量,文中还提出一种双门限邻域平均算法.使0.5bit/pel编码图象的峰峰信噪比提高约3.6dB.

关 键 词:图象压缩  沃尔什-哈达玛变换  活性指数  门限

WHT Image Compression and Enhancement
Cat'''' Dejttn Tang Zfienyu.WHT Image Compression and Enhancement[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,1991(Z1).
Authors:Cat' Dejttn Tang Zfienyu
Institution:Cat' Dejttn Tang Zfienyu
Abstract:A Walsh-Hadamard Transform (WHT) coding for image compression is presented. In the method proposed, the coefficient matrices after orthogonal transformation are divided into three kinds according to the activity index. Then adaptive threshold compression, bit distribution and quantization are carried out. The design principles are discussed and experimental results arc given. In order to improve the quality of the coding image at a low code rate, a neighboring region average-based algorithm for two threshold is proposed. Thus the peak-peak signal-to-noise ratio for 0.5 bit/pel coding images can be increased by about 3.6 dB.
Keywords:Image compression  Walsh-Hadanjard Transform  Activity index  Threshold
本文献已被 CNKI 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号