节点文献

改进对向传播神经网络应用于码书设计及其性能分析

Modified CounterPropagation Network Applied to Codebook Design and Its Performance Analysis

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 佘春东李磊戴彦群王茂芝范植华

【Author】 She Chundong~1) Li Lei~1) Dai Yanqun~2) Wang Maozhi~3,4) Fan Zhihua~1) ~ ~1) (General Software Laboratory, Institute of Software, Chinese Academy of Sciences, Beijing 100080)~ ~2) (School of Engineering Technology, Chengdu University of Technology, Lesan 614007)~ ~3) (School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054)~ ~4) (School of Management of Information, Chengdu University of Technology, Chengdu 610059)

【机构】 中国科学院软件研究所通用软件实验室成都理工大学工程技术学院电子科技大学计算机科学与工程学院中国科学院软件研究所通用软件实验室 北京100080北京100080乐山614007成都610054成都理工大学信息管理学院成都610059北京100080

【摘要】 针对对向传播神经网络(CPN)应用于矢量量化时的两个缺陷进行改进,提出了一种码书设计算法——快速竞争学习及误差修正算法(FCLECA),并设计了相应的基于改进CPN的快速矢量量化器模型,详细讨论了FCLECA的重要步骤、重要参数及其时间复杂度仿真实验结果表明文中算法能在提高码书质量的同时大幅缩短训练时间,是一种有效的快速矢量量化算法

【Abstract】 A new fast codebook designing algorithm referred to as FCLECA (fast competitive learning and error correction algorithm) and based on a modified CPN(counterpropagation network) is presented. The corresponding fast vector quantizer model based on the modified CPN is illustrated. Details of the critical steps and parameters in FCLECA and its time complexity are discussed. Results of simulating experiments indicate that with the new algorithm the training time in codebook designing is greatly reduced, while quality of the generated codebooks is improved. The new algorithm is effective and feasible in vector quantization.

【基金】 中国科学院知识创新工程方向性研究项目基金(KGCX2JG09)
  • 【文献出处】 计算机辅助设计与图形学学报 ,Journal of Computer Aided Design & Computer Graphics , 编辑部邮箱 ,2005年04期
  • 【分类号】TN911.2
  • 【被引频次】2
  • 【下载频次】123
节点文献中: