摘要
手写签名认证是人工智能领域的一个重要课题,对它的深入研究具有重要的现实意义。文章采用基于概率统计的方法表示手写签名的速度特征,用vague集相似度表示书写压力特征,并通过细分的拐点表示笔画特征进行签名认证,最后,构造了一个在线手写签名认证系统。实验证明了该方法的有效性。
The handwritten signature authentication is one of the important issues in artificial intelligence field, and the in-depth study on it has significant practical meaning. The signature authentication is achieved by using the method based on probability statistics to express the speed feature of handwritten signature, the similarity of vague sets to express the pressure feature of writing, and the subdivided turning point to express the stroke feature of characters. Finally, an online handwritten signature authentication system is constructed. The experiments prove the validity of the method.
出处
《计算机时代》
2010年第8期7-9,13,共4页
Computer Era
关键词
手写签名
特征提取
概率统计
VAGUE集
拐点特征
签名真伪判别
handwritten signature
feature extraction
probability statistics
vague sets
turning point feature
signature verification
