| 基于支持向量机的手写签名研究 |
| |
| 引用本文： | 张国富,凌捷,彭辉,谷保平.基于支持向量机的手写签名研究[J].微机发展,2008,18(5):57-59. |
| |
| 作者姓名： | 张国富 凌捷 彭辉 谷保平 |
| |
| 作者单位： | 广东工业大学计算机学院 广东广州510006 |
| |
| 基金项目： | 广东省科技计划项目(2006B11201014,2005B10101067);天河区科技计划项目(061G009) |
| |
| 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。
|
| 关 键 词： | 支持向量机 手写签名 特征提取 |
| 文章编号： | 1673-629X(2008)05-0057-03 |
| 修稿时间： | 2007年8月27日 |
Research of Handwritten Signature Based on SVM |
| |
| ZHANG Guo-fu,LING Jie,PENG Hui,GU Bao-ping.Research of Handwritten Signature Based on SVM[J].Microcomputer Development,2008,18(5):57-59. |
| |
| Authors: | ZHANG Guo-fu LING Jie PENG Hui GU Bao-ping |
| |
| Affiliation: | ZHANG Guo-fu, LING Jie, PENG Hui, GU Bao-ping (Faculty of Computer, Guangdong University of Technology, Guangzhou 510006, China) |
| |
| 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. |
| |
| Keywords: | support vector machine handwritten signature feature selection |
| 本文献已被 CNKI 维普 等数据库收录！ |
