基于改进的支持向量机的脱机中文签名验证

Off-line Chinese signature verification based on the improved support vector machine

采用支持向量机算法来验证脱机中文签名。针对支持向量机算法的不足,将粗糙集和支持向量机相结合,利用粗糙集理论对数据属性进行约简,在某种程度上减少支持向量机求解的计算量。不但避免了特征提取中维数灾问题,还有效改善了训练时间。实验结果表明:粗糙集和支持向量机算法应用于离线签名识别,在相同条件下的识别效果优于支持向量机算法。

By using the support vector machine algorithm of off-line Chinese signature verification.The support vector machine algorithm,rough set and support vector machine combination,using rough set theory to data attributes,to a certain extent reduced support vector machine solving calculation.Not only can the feature extraction in a number of disaster,but also improves the training time.The experimental results show that： the rough set and support vector machine algorithm is applied to off-line signature recognition,in the same condition recognition effect is better than the algorithm of support vector machine.