GLCM和DWT特征在打印文件机源认证中的应用

为了有效解决打印文件机源认证问题,提出了一种基于统计纹理特征选择的打印文件机源认证方法。综合考虑打印字符图像的空间域和时频域特性,将GLCM和DWT统计纹理特征进行组合,运用ReliefF算法实现组合特征的初选,二次特征选择使用SVM-RFE算法。文中实验结果表明,在英文相同字有重复样本集和中文不同字无重复样本集上的分类准确率分别为95.20%和75.00%；特征组合与特征选择有利于提高打印文件机源认证的分类鉴别性能。

Application of GLCM and DWT Features in Source Printer Authentication for Printed Documents

In order to effectively solve the problem of source authentication of printed documents, a method of source printer authentication for print documents based on statistical texture feature selection is proposed. Considering the spatial domain and frequency domain properties of character images, the GLCM and DWT statistical texture features are combined, and the combined statistical texture features are selected twice. First ReliefF algorithm is used to select the combined statistical texture features, and the SVM-RFE feature selection algorithm which based on data learning is choose for the second feature selection. The experimental results show that the classification accuracy rate of the English word sets with duplicate sample and the Chinese word sets without duplicate sample are 95.20% and 75.00% respectively, and feature combination and feature selection are helpful to improve the classification and authentication performance of source printer authentication.