一种基于Retinex和PCA的人脸图像识别方法

摘要

针对目前人脸识别算法在光照条件恶劣时识别精度较低的缺陷,提出一种基于Retinex和PCA的人脸图像识别方法.Retinex算法能够有效去除图像中光照恶劣导致的阴影,而PCA能够有效提取图像中有代表性的特征,从而使得快速准确的识别成为可能.在Yale和Yale B数据库上验证该算法的性能,结果证明,此算法简单快速,且具有较高的识别精度,是一种实用的人脸图像识别方法.%In this paper, a new Retinex and PCA based face recognition method is proposed to improve the recognition accuracy from images with bad illumination. Retinex algorithm can effectively remove the shadow of light lead to bad image illumination, and the PCA can effectively extract the representative features of images, so allowing rapid and accurate identification become possible. Some experiments are taken on Yale and Yale B database to investigate the performance of the algorithm, and the results show that the proposed algorithm is simple, fast, and can achieve high recognition accuracy. Therefore, it is a feasible face recognition method in practical applications.