Handwritten Character Recognition Using Integral-gradient Normalization Method and Directional Feature

摘要：

Character normalization and feature extraction are the key parts in handwritten character recognition system. In this paper,we propose a new normalization method for handwritten character recognition,called integral-gradient normalization based on a new definition of line density. We also improve handwritten character recognition method by using a fast 8-directional pattern character feature vector. A single 8-directional pattern is divided into two different 4-directional patterns. The experiment results demonstrated that integral-gradient normalization had good performance and one improved feature extraction method performed better than most of others.

关键词： handwritten character recognition integral-gradient normalization line density directional feature

作者: Kai Liu Yong Zhao Yule Yuan

作者单位: The Key Laboratory of Integrated Microsystems,Shenzhen Graduate School,Peking University Shenzhen,China

会议类型: 国际会议

会议名称: 2011 International Conference on Opto-Electronics Engineering and Information Science(2011光电电子工程与信息科学国际会议 ICOEIS 2011)

会议地点: 西安

会议语种:英文

页码: 1857-1861

在线出版日期: 2011-12-23（万方平台首次上网日期，不代表论文的发表时间）