计算机工程与应用 ›› 2017, Vol. 53 ›› Issue (20): 105-110.DOI: 10.3778/j.issn.1002-8331.1604-0283

• 模式识别与人工智能 • 上一篇    下一篇

基于模板匹配的移动设备手写汉字笔画识别

洪  洋1,乔晓君2,白晓东3   

  1. 1.南京师范大学 物理科学与技术学院,南京 210023
    2.南京师范大学 教育科学学院,南京 210097
    3.南京师范大学 计算机科学与技术学院,南京 210036
  • 出版日期:2017-10-15 发布日期:2017-10-31

Stroke recognition of Chinese characters handwriting on mobile devices based on template matching

HONG Yang1, QIAO Xiaojun2, BAI Xiaodong3   

  1. 1.School of Physics and Technology, Nanjing Normal University, Nanjing 210023, China
    2.School of Education Science, Nanjing Normal University, Nanjing 210097, China
    3.School of Computer Science and Technology, Nanjing Normal University, Nanjing 210036, China
  • Online:2017-10-15 Published:2017-10-31

摘要: 针对当前移动设备上手写汉字流行的[xml]文件存储格式,提出了一种对用户字笔画与模板字笔画测试匹配的算法,该算法通过方位、拓扑关系和形状3种特征综合量度笔画间的匹配,实验效果良好。该算法在用户字的多笔少笔判别、笔顺的正误性判别、整字的正确性以及工整性判别等方面都有着广泛的应用。

关键词: Hausdorff距离, 拓扑关系相似性量度, 方向链码编码, 总相似度矩阵

Abstract: For the handwritten Chinese mainstream storage format of the mobile equipment, this paper proposes a handwritten Chinese character strokes recognition algorithm between the user strokes and the template strokes. The algorithm is based on the position, topological relation and shape of three kinds of features to measure the matching between the strokes, and the experimental effect is good. The algorithm in user word excess or short of strokes discrimination, stroke identification, the whole character of the correctness and aesthetic judgment, etc. has a wide range of applications.

Key words: Hausdorff distance, topological relation similarity measure, directional chain code, total similarity matrix