基于多特征的支持向量机印鉴识别  

A SVM Classification Algorithm based Multiple Supplemental Features

作　　者：李晗[1] 窦爱萍[2] 解争龙[3] 张琰[2] 

机构地区：[1]长安大学信息工程学院,陕西西安710064 [2]中国航空计算技术研究所,陕西西安710068 [3]咸阳师范学院计算机系,陕西咸阳712000

出　　处：《航空计算技术》2006年第4期32-34,共3页Aeronautical Computing Technique

摘　　要：印鉴真伪鉴别的难点要求识别系统同时具备同类印鉴的鲁棒性及异类印鉴的敏感性。针对这一难点,本文提出了一种基于多特征的支持向量机(SupportVectorM ach ine,SVM)鉴别算法,根据多类特征以及支持向量机的自适应寻优特性,获得对真伪印鉴的鉴别。采用Gabor滤波器获得频率特征,采用差图像获得结构特征,采用原图像和极坐标图像的奇异值获得不变量特征。采用支持向量机(SupportVectorM ach ine,SVM)对印鉴进行真伪鉴别。实验表明,本文方法具有很高的真伪鉴别能力。This paper presents a SVM （Support Vector Machine）classification algorithm based multiple supplemental features extraction to improve the efficiency of seal imprint verification. Genuine and forgery seal imprints are classified according to effective supplement of multiple features. Structural features are extracted from diff - image. Multiple channels Gabor filters are introduced to extract frequency domain features; SVD is applied to polar coordinates image to obtain invariant algebraic features. SVM classifier is adopted to implement final classification decision. Experimental resuits show that high recognition rate can be achieved.

关 键 词：GABOR滤波器 极坐标变换 奇异值分解 支持向量机 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]