基于自适应FCM-NMF的人脸识别研究
|
摘要
非负矩阵分解(NMF)是一种有效提取特征的方法,但算法中参数的随机初始化使得迭代求解速度慢,且易陷入局部极小的问题。针对以上问题,提出了一种自适应FCM-NMF的方法,该方法利用模糊C聚类方法 (FCM)获得相似性关系矩阵,能为NMF参数的初始化提供较好的初值,从而有效解决了上述问题。通过在两个人脸库的实验结果显示,收敛速度明显高于随机赋初值的方法,识别率也有所提高。
Non-negative matrix factorization(NMF) is an effective method to extract features. However the random initialization of parameters in the algorithm makes the iterative solution speed slowly and easy to fall into the local minimal problem. Aiming at the above problem, an adaptive FCM-NMF method is proposed. The fuzzy c-means clustering method(FCM) is used to obtain the similarity relation matrix, which can provide a good initial value for the initialization of NMF parameters, thus effectively solving the above problems. The experimental results on two face libraries show that the convergence rate is significantly higher than the random initial value method, and the recognition rate is also improved.
Non-negative matrix factorization(NMF) is an effective method to extract features. However the random initialization of parameters in the algorithm makes the iterative solution speed slowly and easy to fall into the local minimal problem. Aiming at the above problem, an adaptive FCM-NMF method is proposed. The fuzzy c-means clustering method(FCM) is used to obtain the similarity relation matrix, which can provide a good initial value for the initialization of NMF parameters, thus effectively solving the above problems. The experimental results on two face libraries show that the convergence rate is significantly higher than the random initial value method, and the recognition rate is also improved.
引文
[1] NAZMEEN B B,SUNILDUTH B. Fusion of local and global features for face recognition[C]//2015 International Conference on Computing,Communication and Security(ICCCS). Mauritius:IEEE Conference Publications,2015:1-8.
[2] EYAD I A,MOHAMMED E S,KHALIDA S R. Face recognition rate using different classifier methods based on PCA[C]//2017 International Conference on Current Research in Computer Science and Information Technology(ICCIT). Lebanon:IEEE Conference Publications,2017:37-40.
[3] HUWEDI A S,SELEM H M. Face recognition using regularized linear discriminant analysis under occlusions and illumination variations[C]//20164th International Conference on Control Engineering&Information Technology(CEIT),Hammamet,Tunisia,2016:1-5.
[4] ZE L,XU D J,ALEX K. A novel LBP-based color descriptor for face recognition,2017[C]//2017 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP). Shanghai:IEEE Conference Publications,2017:1857-1861.
[5] LEE D D,SEBASTIAN S H. Learning the parts of objects by non negative matrix factorization[J]. Nature,1999,401(6755):788-791.
[6] LEE D D,SEBASTIAN S H. Algorithms for non negative matrix factorization[C]//Neural Information Processing Systems. 2000:556-562.
[7] LI S,HOU X W,ZHANG H J,et al. Learning spatially localized parts-bases representation[C]//Computer Vision and Pattern Recognition. 2001:1063-6919.
[8] WANG Y,JIA Y D,HU C B,et al. Fisher Non-negative Matrix Factorization for learning local features[C]//In Proc Asian Conf On Comp Vision.2004:27-30.
[9] CAI D,HE X F,WU X Y,et al. Non-negative matrix factorization on manifold[C]//IEEE International Conference on Data Mining. 2008:63-72.
[10]刘昱昊.基于基于非负矩阵分解算法的人脸识别技术的研究[D].吉林:吉林大学. 2014.
[11]杨宏伟.基于非负矩阵分解的人脸识别研究[D].兰州:西北师范大学. 2015.
[12] LI B F,TANG Y D,HAN Z. Research on face recognition algorithm based on improved non negative matrix factorization[J]. Computer Simulation,2016,33(3):428-432.
[13] CHEN W S,ZHAN Y,PAN B,et al. Supervised kernel nonnegative matrix factorization for face recognition[J]. Neurocomputing,2016,205:165-181.
[14] BEZDEL J C. Pattern recognition with fuzzy objective function algorithms[M]. New York:Plenum,1981.
[15] WANG W N,ZHANG Y J,et al. 2006 6th World Congress on Intelligent Control and Automation:The Global Fuzzy c-means Clustering Algorithm,2006[C]//Dalian:IEEE Conference Publications,2006:3604-3607.
[16] ZARINBAL M,FAZEL Z M H,TURKSEN I B. Relative entropy fuzzy c-means clustering[J]. Information Sciences,2014,260:74-97.
[17]钟毅.一种基于相关系数的模糊C均值聚类算法[J].软件产业与工程,2016(6):50-53.
[18] PAL N R,BEZDEK J C. On cluster validity for the fuzzy C-means model[J]. IEEE Transactions on Fuzzy Systems,1995,3(3):370-379.
[2] EYAD I A,MOHAMMED E S,KHALIDA S R. Face recognition rate using different classifier methods based on PCA[C]//2017 International Conference on Current Research in Computer Science and Information Technology(ICCIT). Lebanon:IEEE Conference Publications,2017:37-40.
[3] HUWEDI A S,SELEM H M. Face recognition using regularized linear discriminant analysis under occlusions and illumination variations[C]//20164th International Conference on Control Engineering&Information Technology(CEIT),Hammamet,Tunisia,2016:1-5.
[4] ZE L,XU D J,ALEX K. A novel LBP-based color descriptor for face recognition,2017[C]//2017 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP). Shanghai:IEEE Conference Publications,2017:1857-1861.
[5] LEE D D,SEBASTIAN S H. Learning the parts of objects by non negative matrix factorization[J]. Nature,1999,401(6755):788-791.
[6] LEE D D,SEBASTIAN S H. Algorithms for non negative matrix factorization[C]//Neural Information Processing Systems. 2000:556-562.
[7] LI S,HOU X W,ZHANG H J,et al. Learning spatially localized parts-bases representation[C]//Computer Vision and Pattern Recognition. 2001:1063-6919.
[8] WANG Y,JIA Y D,HU C B,et al. Fisher Non-negative Matrix Factorization for learning local features[C]//In Proc Asian Conf On Comp Vision.2004:27-30.
[9] CAI D,HE X F,WU X Y,et al. Non-negative matrix factorization on manifold[C]//IEEE International Conference on Data Mining. 2008:63-72.
[10]刘昱昊.基于基于非负矩阵分解算法的人脸识别技术的研究[D].吉林:吉林大学. 2014.
[11]杨宏伟.基于非负矩阵分解的人脸识别研究[D].兰州:西北师范大学. 2015.
[12] LI B F,TANG Y D,HAN Z. Research on face recognition algorithm based on improved non negative matrix factorization[J]. Computer Simulation,2016,33(3):428-432.
[13] CHEN W S,ZHAN Y,PAN B,et al. Supervised kernel nonnegative matrix factorization for face recognition[J]. Neurocomputing,2016,205:165-181.
[14] BEZDEL J C. Pattern recognition with fuzzy objective function algorithms[M]. New York:Plenum,1981.
[15] WANG W N,ZHANG Y J,et al. 2006 6th World Congress on Intelligent Control and Automation:The Global Fuzzy c-means Clustering Algorithm,2006[C]//Dalian:IEEE Conference Publications,2006:3604-3607.
[16] ZARINBAL M,FAZEL Z M H,TURKSEN I B. Relative entropy fuzzy c-means clustering[J]. Information Sciences,2014,260:74-97.
[17]钟毅.一种基于相关系数的模糊C均值聚类算法[J].软件产业与工程,2016(6):50-53.
[18] PAL N R,BEZDEK J C. On cluster validity for the fuzzy C-means model[J]. IEEE Transactions on Fuzzy Systems,1995,3(3):370-379.