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

一种改进的等范数最近邻码本矢量搜索算法
引用本文:蒋建国,彭兴邦,聂秋玉,苌凝凝.一种改进的等范数最近邻码本矢量搜索算法[J].合肥工业大学学报(自然科学版),2007,30(1).
作者姓名:蒋建国  彭兴邦  聂秋玉  苌凝凝
作者单位:合肥工业大学,计算机与信息学院,安徽,合肥,230009
基金项目:国家自然科学基金 , 安徽省二期科技攻关重大计划
摘    要:文章提出了一种改进的等范数最近邻码本矢量搜索算法,该算法预先离线计算好各码本矢量的2-范数和它与中心线的矢量夹角,再根据各码本矢量的2-范数进行升序排列.编码时用二分法搜索到与目标矢量2-范数最近的码本矢量,再排除那些不在目标矢量形成的夹角范围内的码本矢量。结果证明,该算法对比文中列举的其他算法更加高效。

关 键 词:矢量量化  矢量夹角  码字搜索

An improved equal-norm nearest neighbor codeword search algorithm for vector quantization
JIANG Jian-guo,PENG Xing-bang,NIE Qiu-yu,CHANG Ning-ning.An improved equal-norm nearest neighbor codeword search algorithm for vector quantization[J].Journal of Hefei University of Technology(Natural Science),2007,30(1).
Authors:JIANG Jian-guo  PENG Xing-bang  NIE Qiu-yu  CHANG Ning-ning
Abstract:An improved codeword searching algorithm is proposed on the basis of the equal-norm nearest neighbor search algorithm.The norm and the angle made with the center line by the vector of the codeword are computed for each codeword off-line firstly,then the codewords are ordered according to the ascending order of their norm.During the encoding process,the algorithm first selects the codeword that has the minimum norm distance from the input vector as the tentative matching codeword,and then removes the codeword which does not fall into the angle region formed by the input vector.Experimental results show that this algorithm is more efficient than other common algorithms.
Keywords:vector quantization  vector angle  codeword search
本文献已被 CNKI 万方数据 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