The Chinese character is made up of strokes or sub-strokes. A fuzzy sub-stroke extraction method is proposed to resolve the unsteadiness because of the unconstrained written fashion.
在特征抽取与分类识别方面，本文提出一种模糊子笔画抽取方法，解决了因无限制手写体笔画随意性而使得抽取的子笔画不稳定的问题。
参考来源 - 小类别无限制手写体汉字识别研究·2,447,543篇论文数据，部分数据来源于NoteExpress
银行支票手写体汉字大写金额识别的实验结果表明应用模糊子笔画统计特征能取得更好的识别效果。
Experiments of handwritten legal amount recognition on Chinese check show that the fuzzy sub-stroke statistical feature proposed in this paper is more effective.
论文提出一种模糊子笔画抽取方法，解决了因无限制手写体笔画随意性而使得抽取的子笔画不稳定的问题。
A fuzzy sub-stroke extraction method is proposed to resolve the unsteadiness because of the unconstrained written fashion.
然后利用图模型来对子笔画和模糊区域进行建模，同时通过构造贝叶斯分类器来分析子笔画对的连续性，并通过路径搜索来得到子笔画序列；
Then, the sub-strokes and ambiguous zones can be modeled with a graph, and a Bayesian classifier is built to analyze the continuity of sub-stroke pairs.
应用推荐