基于幂次变换预处理的PCA人脸识别算法

摘要

This paper proposed an efficient method for face recognition. It developed a pretreatment strategy that could deal with image gradation to restrain upper sensitivity of PCA from the change of illumination in an image. The training and testing samples were selected with random sequences from the face database, processed with power transformation and a Butterworth low-pass filter, and finally processed with the PCA algorithm. The experimental results based on the ORL database show that the achieved recognition rate with the proper power transformation values is 96. 70%. Therefore, the proposed algorithm can have higher accuracy of recognition than the traditional ones.%为了抑制主成分分析(PCA)对图像中光照等变化的较高敏感性、进一步提高人脸识别率,提出了一种对图像灰度进行幂次变换预处理的策略.首先采用随机序列来选取人脸库中的训练样本和测试样本,然后对随机人脸样本进行幂次变换和Butterworth低通滤波处理,最后应用PCA处理的人脸识别算法.基于ORL数据库的实验表明,在适当选择幂次变换参数的情况下,基于幂次变换预处理的PCA人脸识别算法平均人脸识别率达到96.70％.因此,基于幂次变换预处理的PCA人脸识别算法比传统的PCA算法具有更高的识别精度.