机构地区: 广东工贸职业技术学院计算机工程系
出 处: 《计算机与现代化》 2013年第11期210-213,共4页
摘 要: 基于人脸图像的年龄分类是指将人脸图像分为预定义的几个年龄组的方法,它是一个多分类问题。提出一种融合纠错输出编码的SVM多类分类器,将二分类器SVM扩展为多分类器。采用Gabor小波提取人脸年龄特征,并应用二元主成分(2DPCA)分析法对提取的特征进行降维,在FG-NET年龄数据库上进行实验,结果证明了该方法的有效性和鲁棒性。在人脸年龄特征提取方面,Gabor与2DPCA结合的方式比单纯2DPCA方式具有更好的年龄特征表达能力。 Face image-based age classification is an approach to classify face images into one of several pre-defined age-groups, it is a muhiclass problem. For solving the multiclass problem, a SVM extension is proposed which combines the method of error- correcting output codes (ECOC) with binary SVM classifier. Gabor aging features are extracted to represent the face images, and the two principal components (2DPCA) are used to reduce the dimension of the extracted features. Experimental results on FG- NET database are reported to demonstrate its effectiveness and robustness. The results obtained using the fused Gabor and 2DPCA features are better than the one when using 2DPCA alone.
领 域: [自动化与计算机技术] [自动化与计算机技术]