基于空间曲线弹性匹配的在线手写签名鉴别  

On Line Handwriting Signature Verification Based on Elastic Matching of 3D Curve

作　　者：金涌[1] 柳健[1] 

机构地区：[1]华中理工大学图像识别与人工智能研究所图像信息处理与智能控制国家教委开放研究实验室

出　　处：《华中理工大学学报》1999年第5期14-16,共3页Journal of Huazhong University of Science and Technology

基　　金：国防科技预研跨行业基金

摘　　要：论述了将手写签名笔划轨迹视为空间曲线，并采用点－点加权弹性匹配对其进行鉴别的方法．该方法强调签名在形状和签名压力的动态特征上都具有相似性．算法中权重系数根据签名样本的稳定度来确定，加权的匹配算法考虑了局部相似性不同的特点，有利于认证指标的改善．A handwriting signature verification method was introduced based on point to point weighted elastic matching. The handwriting signature strokes were considered as 3D curves in space. This approach emphasizes both static feature of signature shape and dynamic feature of tip pressure. The weight coefficient was decided by the stability of signature samples in the algorithm.

关 键 词：手写签名鉴别 弹性匹配 空间曲线匹配 点-点匹配 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]