摘要
为了克服传统LBG算法中的空胞腔现象,提出了一种基于码字间距最大化的新的空胞腔策略。利用离码书距离最大的输入矢量来修改胞腔中的码字,旨在形成码字的合理分布,减小矢量量化的平均失真。实验结果表明:提出的策略能有效地消除空胞腔现象,获得性能较好的码书,其峰值信噪比比传统的LBG算法提高了3dB。
To tackle the empty cell phenomenon in the LBG algorithm,this paper presents a new empty cell strategy which is based on maximization the distance between the code words.The empty cell is filled with the input vector which has the largest distance to the codebook.The goal is to get an even distribution of the code words,alleviating the overall distorting in vector quantization.Experimental results show:the presented strategy can resolve the empty ceils problem effectively,outperforming the standard LBG and other competing algorithms.The improvement of peak signal-noise ratio is 3 dB compared with the conventional LBG algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第28期40-41,112,共3页
Computer Engineering and Applications
基金
国家自然科学基金No.60673191
广东外语外贸大学创新基金资助项目( No.GW2006- TB- 012)~~
关键词
图像压缩
矢量量化
LBG算法
空胞腔
间距最大化
image compression
vector quantization
LBG algorithm
empty cells
maximization the distance