基于奇异值特征的图像预处理及人脸识别

IMAGE PREPROCESSING BASED ON SINGULAR VAILUE AND FACE RECOGNITION

从增强图像的分类信息角度出发，对人脸图像做频域上的带通滤波预处理，来提高奇异值特征的分类性能．给出了一组圆形滤波函数，并使用遗传算法来选择可分性较大的频段，设计了染色体的表示方式和适应度的求法．

In order to enhance the classification information of a image, we use band pass filtering to preprocess face image and to improve the sort performance of singular value feature. We give a group of circular filter functions, use genetic algorithm to select the frequency band of larger divisibility value, and design the expression of chromosome and the solution of fitness. Experiments prove that we can get better sort result of singular value feature by doing preprocessing than undoing it.