基于基本笔画起收笔特征的手写汉字的笔迹鉴定

作者基于汉字基本笔画起收笔中蕴含丰富的个人特征的设想,提出以手写汉字基本笔画为研究对象,抽取其中起收笔特征,形成特征空间,求得鉴定对象在特征空间中的分布,并计算它与辞书中各书写者之间的广义距离,从而实现笔迹鉴定.该研究选用10位书写者,4种基本笔画,每种基本笔画采用10个样本进行实验.结果证实此方法完全克服了以往笔迹鉴定研究中结体依存性带来的不足,只需对少数基本笔画的研究就对全体汉字有效,进一步验证了基于基本笔画的笔迹鉴定研究方法的可行性和有效性.

Identification of Hand-Written Chinese Characters Based on Features of the First and Last Strokes

This research is based on an assumption that the first and last strokes of a hand-written Chinese character contain rich personal features. Taking the basic strokes as main objects, features of the first and last strokes are used to form a multi-dimensional feature space, and the distribution of individual writers in the feature space is obtained. By calculating the distances between the feature of a specific writer and that of the other writers in the dictionary, the writer identification can be found. In the research, ten writers are selected. Each writer provides 10 samples for each of the four basic strokes. This overcomes the drawback of text-dependence in the previous research, and further verifies the feasibility and validity of the method based on basic strokes.