| 改进的基于主分量分析的码书设计算法 |
| |
| 引用本文： | 孙宏伟,顾明,孙家广. 改进的基于主分量分析的码书设计算法[J]. 计算机辅助设计与图形学学报, 2005, 17(10): 2245-2250 |
| |
| 作者姓名： | 孙宏伟 顾明 孙家广 |
| |
| 作者单位： | 清华大学计算机科学与技术系,北京,100084;清华大学软件学院,北京,100084;清华大学软件学院,北京,100084 |
| |
| 基金项目： | 国家重点基础研究发展规划项目（2004CB719400）;国家“八六三”高技术研究发展计划（2003AA413030,2003AA414030） |
| |
| 摘 要： | 针对基于主分量分析和遗传算法的码书设计算法中当码书大小超过64时码书性能下降的问题，提出了一种改进的码书设计算法．首先采用主分量分析对训练矢量降维以减少计算复杂度，然后利用遗传算法的全局优化能力计算得到接近全局最优的码书．实验结果表明，与原算法和经典的LBG算法相比，文中算法所生成的码书性能有了明显提高，而且计算时间也少于LBG算法．
|
| 关 键 词： | 矢量量化 码书设计 主分量分析 遗传算法 |
| 收稿时间： | 2004-07-15 |
| 修稿时间： | 2004-07-152005-05-10 |
Improved Codebook Design Algorithm Based on Principal Component Analysis |
| |
| Sun Hongwei,Gu Ming,Sun Jiaguang. Improved Codebook Design Algorithm Based on Principal Component Analysis[J]. Journal of Computer-Aided Design & Computer Graphics, 2005, 17(10): 2245-2250 |
| |
| Authors: | Sun Hongwei Gu Ming Sun Jiaguang |
| |
| Affiliation: | 1 Department of Computer Science and Technology, Tsinghua University, Beijing 100084; 2 School of Software, Tsinghua University, Beijing 100084 |
| |
| Abstract: | This paper proposes an improved codebook design algorithm for image compression, which solves the problem that the codebook design algorithm based on principal component analysis and genetic algorithm obtains the worse codebooks than those computed by LBG algorithm when the codebook size is larger than 64. By the algorithm, we first reduce the dimensionality of the training vectors using principal component analysis, and then use the near global optimal searching ability of genetic algorithm to compute the codebook. Experimental results show that our algorithm outperforms the primary algorithm and the popular LBG algorithm in terms of image compression performance, and the computing time needed by our algorithm is also shorter than the LBG algorithm. |
| |
| Keywords: | vector quantization codebook design principal component analysis genetic algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
