【摘要】 针对一般手写签名中特征提取方法的不足,将支持向量机的原理引入到手写签名算法里,从而可以很好地应用于高维数据,避免了特征提取中维数灾问题。主要研究如何在标准的窗格中利用扫描的方法提取图像密度特征,从而得到特征向量。通过MATLAB工具,将得到的图像密度特征作为特征向量为SVM的输入进行训练仿真实验。实验表明,该方法能够有效识别手写签名真伪,说明把支持向量机应用到手写签名具有很好的识别能力,并解决了"维数灾"的问题。更多还原

【Abstract】 Not content to method of feature selection applied in the field of handwritten signature,the theory of support vector machine applied in the algorithm which better applied in high-dimension data avoid the question of dimension-destroy in the feature selection.Recognition application of handwritten signature based on the principle and application of support vector machine is proposed.The background and development prospects are described.Through testing based on the MATLAB tool,support vector machine has good ability for recognition in the handwritten.The result shows that the proposed method has encouraging performance.更多还原