| 基于改进SOFM的矢量量化图像压缩 |
| |
| 引用本文： | 王茂芝,郭彬,徐文皙.基于改进SOFM的矢量量化图像压缩[J].成都理工大学学报(自然科学版),2007,34(6):648-652. |
| |
| 作者姓名： | 王茂芝 郭彬 徐文皙 |
| |
| 作者单位： | 成都理工大学信息管理学院,成都,610059;电子科技大学计算机学院,成都,610059 |
| |
| 基金项目： | 数学地质四川省高校重点实验室资助项目
,
四川省教育厅资助项目
,
成都理工大学校科研和教改项目 |
| |
| 摘 要： | 在介绍矢量量化和自组织特征映射神经网络的基础上,针对基于自组织特征映射神经网络的矢量量化算法,在初始码书生成、获胜神经元搜索策略以及调整获胜码字及其拓扑领域权值等方面进行改进.实验结果表明改进算法具有合理性和有效性.
|
| 关 键 词： | 自组织特征映射 矢量量化 码书 图像压缩 |
| 文章编号： | 1671-9727(2007)06-0648-05 |
| 修稿时间： | 2007年2月12日 |
Image impression method of vector quantization based on the improved SOFM |
| |
| WANG Mao-zhi,XU Wen-xi.College of Information Management,Chengdu University of Technology,Chengdu ,China,.School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu ,China.Image impression method of vector quantization based on the improved SOFM[J].Journal of Chengdu University of Technology: Sci & Technol Ed,2007,34(6):648-652. |
| |
| Authors: | WANG Mao-zhi XU Wen-xiCollege of Information Management Chengdu University of Technology Chengdu China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China |
| |
| Institution: | WANG Mao-zhi1,XU Wen-xi11.College of Information Management,Chengdu University of Technology,Chengdu 610059,China,2.School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu 610054,China |
| |
| Abstract: | The theories of vector quantization(VQ) and self-organizing feature mapping(SOFM) neural networks are introduced in this paper firstly.Three aspects are improved based on VQ of SOFM.The first is to generate the initial codebook.The second is to search for the strategy of the winner neuron.And the third is to adjust winner code word and topology field weight.The results of the experiment illustrate the rationality and efficiency of the improved algorithm. |
| |
| Keywords: | self-organizing feature mapping neural networks vector quantization codebook image compression |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
