基于自组织映射神经网络的粘连字符的分割
Merged-symbol Segment Based On Self-organizing Feature Map Neural Networks
用自组织映射神经网络作为粘连字符分割的方法，对经典的自组织学习规则做了一些改进，使其以较少的神经元结点、较快的速度逼近粘连字符的白像素点的分布。
In this paper, a new method, self-organizing feature map is proposed and the classic updating rule of self-organizing feature map is modified to segment merged-symbol. we obtained a network that can approximate the distribution of white-pixels between two symbols in less training time and with less units.
张玉林
计算技术、计算机技术
模式识别粘连字符分割自组织映射
Pattern organizationmerged-symbol segmentself-organizing feature map
张玉林.基于自组织映射神经网络的粘连字符的分割[EB/OL].(2005-10-27)[2025-08-18].http://www.paper.edu.cn/releasepaper/content/200510-288.点此复制
评论