We design the function which describe the attribute probability of the stroke segment. We also define the contributive value of the grid and propose the concept of the expanded grid, and we got good results of Cross Recognition.
我们用网格化特征记录了汉字的结构信息，设计了定义笔段方向隶属度函数、网格的贡献度及扩充网格的概念，有效的反映了汉字的信息。
参考来源 - 基于实虚笔段方向特征粗分方法的联机手写汉字识别·2,447,543篇论文数据，部分数据来源于NoteExpress
A stroke segment-based algorithm was proposed, which is equal to the problem of ordering the stroke segments in nature.
基于笔段的笔迹重构方法中，笔迹重构实质上就是笔段的排序问题。
An algorithm for stroke segment extraction directly from Chinese character dot matrix is presented. It avoids the process of thining.
本文提出了从汉字点阵中直接抽取笔划特征的新算法，省去了细化过程。
This paper takes stroke segment as pattern primitive to represent the structure information of Chinese character. On the basis, a new algorithm to recognize handwritten Chinese character is proposed.
以笔划为基元结合笔划的顺序来表示汉字的结构信息，在此基础上提出了一种手写汉字识别的匹配算法。
应用推荐