| 用改进的竞争Hopfield神经网络求解多边形近似问题 |
| |
| 引用本文： | 旷章辉,王甲海,周雅兰.用改进的竞争Hopfield神经网络求解多边形近似问题[J].计算机科学,2009,36(3). |
| |
| 作者姓名： | 旷章辉 王甲海 周雅兰 |
| |
| 作者单位： | 1. 中山大学信息科学与技术学院,广州,510006
2. 广东商学院信息学院,广州,510320 |
| |
| 基金项目： | 广东省自然科学基金，高等学校博士学科点专项科研基金，教育部留学回国人员科研启动基金 |
| |
| 摘 要： | 多边形近似是提取曲线特征点和简化曲线描述的一种重要方法.提出一种改进的Hopfield神经网络多边形近似算法,该算法利用选择拐点策略减少了搜索空间,重新定义了神经网络的能量函数,使其更能反映优化目标;引?入合并拆分搜索策略,有效帮助神经网络脱离局部最小值.实验结果表明,提出的改进算法是有效的,比其它算法如关键点检测法、竞争Hopfield神经网络、混沌Hopfield神经网络、遗传算法等具有更优的性能.
|
| 关 键 词： | 竞争Hopfield神经网络 合并拆分 多边形近似 |
Improved Competitive Hopfield Neural Network for Polygonal Approximation Problem |
| |
| KUANG Zhang-hui,WANG Jia-hai,ZHOU Ya-lan.Improved Competitive Hopfield Neural Network for Polygonal Approximation Problem[J].Computer Science,2009,36(3). |
| |
| Authors: | KUANG Zhang-hui WANG Jia-hai ZHOU Ya-lan |
| |
| Affiliation: | School of Information Science and Technology;Sun Yat-Sen University;Guangzhou 510006;China;Information Science School;Guangdong University of Business Studies;Guangzhou 510320;China |
| |
| Abstract: | Polygonal approximation is an important method to detect dominant points and represent curves.An Improved Competitive Hopfield Neural Network (ICHNN) was proposed for polygonal approximation problem.The proposed algorithm reduces research spaces by deleting the collinear points,redefines energy function of the network to represent the aim more efficiently,and combines with a merge-split method to help the network escape from the local mini-ma.The experimental results show that ICHNN is robuster and better t... |
| |
| Keywords: | Improved competitive Hopfield neural network Merge and split Polygonal approximation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
