摘要 提出一种基于支持向量数据描述方法的在线签名身份认证系统.首先,采用双向后向合并DTW(Dynamic Time Warping)算法确定签名中关键点之间的对应关系,然后采用经典DTW度量签名局部中各种细微的差异.文中提出基于差异值均值方差最小原则的特征选择方法.最后,采用支持向量数据描述方法设计分类器.为得到更好的认证效果,采用多层交叉验证和遗传算法寻找最优的分类器参数.在SVC2004数据库上,系统对熟练伪造签名取得4.25%的平均等错误率.
Abstract:An novel online signature verification system is proposed based on support vector data description (SVDD). Firstly, correspondences of the critical points in signatures are confirmed by bidirectional backwardmerging dynamic time wrapping algorithm. Then, subtle differences in the local are calculated by classical dynamic time wrapping algorithm. Feature selection principle based on mean and deviation minimization is proposed. Finally, the classifiers are designed using support vector data description (SVDD). To obtain better result, m-fold-cross validation and genetic algorithm are used to seek optimal parameters of SVDD. The average equal error rate for skill forge signatures in SVC2004 signatures database is 4.25%.
邹杰, 吴仲城. 基于支持向量数据描述的在线签名认证系统[J]. 模式识别与人工智能, 2011, 24(2): 284-290.
ZOU Jie, WU Zhong-Cheng. Online Signature Verification System Based on Support Vector Data Description. , 2011, 24(2): 284-290.