基于极角特征匹配的动态签名鉴别算法  

Dynamic Signature Verification Algorithm Based on Polar Angle Matching

作　　者：程开东[1] 董险峰[1] 马驷良[1] 

机构地区：[1]吉林大学数学学院,长春130012

出　　处：《吉林大学学报（理学版）》2009年第5期1006-1012,共7页Journal of Jilin University:Science Edition

基　　金：吉林省科技发展计划项目基金(批准号:20050327_120050327_2)

摘　　要：基于签名质心和黄金分割质心建立极坐标系,并提取签名极角特征中相对稳定的极值点序列作为笔段分隔点给出一种签名鉴别算法.提出了改进的动态时间规整(DTW)算法,通过该算法将待测样本和签名模板依分隔点的对应关系按段进行动态匹配,得到最优匹配方案和最佳相似度,从而降低了误拒率.算法经JLU-DHSDB2.0签名数据库测试,识别等错率达到4.25%.A method is presented to create a polar coordinate system based on the mass center of whole signature and the mass center of golden section under which the extrema can be extracted from polar angle feature as signature segmentation-point sequence. This article proposes an improved DTW algorithm for on-line handwritten signature verification that is applied to the segments matching between the signature sample and the template automatically, obtaining the highest qualified corresponding relationship of the segments and the most closely similarity measurement result, which can degrade the FRR level efficiently. The algorithm has been tested via the signature data from the handwritten signature database JLU-DHSDB2.0 and the experiment results show that the EER is 4.25 percent.

关 键 词：在线手写签名鉴别 动态时间规整 模式识别 手写签名数据库 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]