基于小波和稀疏矩阵的人脸识别算法  

Research on face recognition algorithm based on wavelet and sparse matrix

作　　者：王国权 乔琪 赵艳芹 巩燕 Wang Guoquan;Qiao Qi;Zhao Yanqin;Gong Yan(School of Computer ＆ Information Engineering,Heilongjiang University of Science ＆ Technology,Harbin 150022,China)

机构地区：[1]黑龙江科技大学计算机与信息工程学院,哈尔滨150022

出　　处：《黑龙江科技大学学报》2018年第4期419-424,共6页Journal of Heilongjiang University of Science And Technology

基　　金：黑龙江省自然科学基金项目(F201436)

摘　　要：为提高光照变化下的人脸识别率,提出一种基于小波和稀疏矩阵的人脸识别算法。利用近似轴对称人脸预处理方法中的梯度下降过程,减小左右人脸图像之间由光照阴影带来的差异,提取人脸图像的小波光照不变量,应用稀疏表示分类器识别人脸光照不变特征,采用四种算法对比分析Extended Yale B和CMU_PIE人脸数据库中人脸识别的准确率。结果表明,文中算法能有效消除光照阴影对小波光照不变量产生的伪轮廓影响,较其他三种算法有较好的识别准确率。该算法在光照变化条件下具有较好的人脸识别效果,可以为人脸准确识别提供参考。This paper proposes a novel face recognition algorithm based on wavelet and sparse matrix in an eefort to improve the face recognition rat^ under illumination changes. The of reducing the differences betweenthe left and right face images due to light shading using the gradiee descent process in the approximate axisymmetric face preprocessing method; extracting the wavelet illumination invarants of the face image, and thereby recognizing the face illumination invarant features using the sparse representation classifier; and ultimately comparng and analyzing the accuracy of face recognition in the Extended Yale B and CMU_PIE1 face databases using four kinds of demonstrate that the proposed algorithm capable of effectively eliminating the effect of light shading on the false contour generated by wavelet illumination invarants boasts a better recognition accuracy than the other three algorithms. The algorithm with a better face recognition effect under the different illumination conditions may provide a reference for an accurate face recognition.

关 键 词：人脸识别 小波光照不变量 近似轴对称人脸预处理 稀疏表示 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]