| 一种基于回归神经网络的非线性矢量量化方法 |
| |
| 引用本文： | 马霓,韦岗. 一种基于回归神经网络的非线性矢量量化方法[J]. 信号处理, 2001, 17(1): 42-46 |
| |
| 作者姓名： | 马霓 韦岗 |
| |
| 作者单位： | 1. 华南理工大学电子信息学院
2. 广州金鹏集团有限公司 |
| |
| 基金项目： | 国家自然科学基金(69602002,69772027)和广东省自然科学基金(960227,963037)资助项目 |
| |
| 摘 要： | 本文提出了一种两级神经网络矢量量化(NNVQ)的方法.该方案中,通过用激励矢量所驱动的网络输出来逼近所要编码的矢量.网络和激励码本都通过优化从而能克服常规算法中的一些病态情况.实验结果表明使用了回归神经网络和优化训练算法的NNVQ的性能好于一般的矢量量化算法.
|
| 关 键 词： | 矢量量化 回归神经网络 训练算法 优化 |
A Vector Quantization by Means of Recurrent neural Networks |
| |
| MA Ni,Wei Gang. A Vector Quantization by Means of Recurrent neural Networks[J]. Signal Processing(China), 2001, 17(1): 42-46 |
| |
| Authors: | MA Ni Wei Gang |
| |
| Abstract: | A vector quantiztion scheme with a two-stage neural network coding (NNVQ) is developed, where an encoded vector is approximated by the output of the networks driven by a vector from an excitation codebook. The networks and the codebook are optimized to overcome some difficulties in the conventional algorithms. The experimental results show that the NNVQ, which employs the recurrent neural networks and the optimized training algorithm, performs the best among the reference versions of the VQ algorithms. |
| |
| Keywords: | Vector quantization Recurrent neural networks Training algorithm Optimization |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
