融合小波变换和张量PCA的人脸识别算法

张量主成分分析(PCA)方法用于人脸识别能获得比PCA方法更高的识别率.小波变换具有良好的时频分析特性,同时还能起到降维的作用.综合利用这两个算法的优点,提出了一种新的人脸识别算法,对人脸图像先采用小波变换做预处理得到4个子带图像,然后对每个子带图像用张量PCA进行特征提取,实现人脸图像的高效识别.仿真结果表明,新算法的识别率比张量PCA方法提高了6%,识别时间为张量PCA方法的35.74%.

Face recognition algorithm based on wavelet preprocessing and tensor PCA

The accuracy rate of the face recognition by tensor PCA is higher than that by PCA. And wavelet has two abilities to capture localized time-frequency information and to reduce the dimension of images. According to the two advantages of the above algorithms, a new face recognition algorithm based on wavelet transform and tensor PCA is proposed. Wavelet transform is firstly used and then tensor PCA is used to extract the feature of subband images, and the efficient recognition of face images can be realized. The recognition rate of the proposed alogorithm is 6％ higher than that of the tensror PCA algorithm, and the recognition time of the proposed algorithm is 35.74％ that of the tensor PCA algorithm, which is illustraed in experimental results.