基于支持向量网络和相关反馈的人脸识别

采用了一种通过KPCA提取人脸图像特征,线性SVM对特征进行加权,用最近邻法分类人脸的识别系统.整个系统实质上构成了一个支持向量分类网络.为了自动进行网络训练和参数寻优,提出了一套自动相关反馈训练方法;并采用了图像灰度的伽马校正技术减少光照变化对识别的影响,提高了分类器的性能.基于ORL数据库的相关实验表明,在很少样本训练条件下,这样的系统能够获得较高性能.

Face recognition based on support vector network and relevance feedback

This paper presented a stratergy for face recognition using KPCA and SVM. The total system was a support vector network for classification task actually. In order to train this network automatically, relevance feedback was utilized for adjusting parameters and a remapping technique was adopted to overcome the illumination problem. These schemes enhanced the performance of this method compare to the traditional PCA and SVM method. The experimental results show that the accuracy of face recognition can be increased with less samples training.