截断核范数的矩阵回归及在人脸识别中的应用
Matrix Regression Based on Truncated Nuclear Norm and Application in Face Recognition
DOI: 10.12677/CSA.2021.113076, PDF,   
作者: 穆 松：南京信息职业技术学院 士官学院，江苏 南京；张治斌：北京信息职业技术学院软件与信息学院，北京
关键词: 稳健回归；截断核范数；交替方向乘子算法；人脸识别；Robust Regression； Truncated Nuclear Norm； ADMM； Face Recognition
摘要: 采用回归方法进行人脸识别的过程中对误差的度量采用像素层的F范数，此方法需要假设像素之间是相互独立的，然而在具有连续遮挡的情况下该假设是不成立的。事实上，误差图像在空间上是相关的，采用矩阵的截断核范数可以更好地描述图像的结构信息。对误差图像进行分析后，提出了一种截断核范数的回归模型，并采用交替方向乘子算法(ADMM)求解。与现有的其他回归方法相比，本文的方法将误差检测和误差矫正集成到一个回归模型中，对Extend Yale B人脸数据库的实验也证明了该方法在人脸识别上的优越性。
Abstract: In the process of face recognition using regression method, the error is measured by the F-norm. This method needs to assume that the pixels are independent of each other, but this assumption is not true in the case of continuous occlusion. In fact, the error images are spatially related, and the truncated nuclear norm of the matrix can describe the structural information of the image. By analyzing the error image, a regression model based on truncated nuclear norm is proposed and solved by the Alternating Direction Multiplier Algorithm (ADMM). Compared with other existing regression methods, the method of this paper integrates error detection and error correction into a regression model. The experiment on Extend Yale B and AR face database also proves the superiority of this method in face recognition.
文章引用：穆松, 张治斌. 截断核范数的矩阵回归及在人脸识别中的应用[J]. 计算机科学与应用, 2021, 11(3): 741-750. https://doi.org/10.12677/CSA.2021.113076

参考文献

[1] CHEN and BLEDSOE (1965) Automatic Face Recognition. Panoramic Research Inc., Palo Alto.
[2] Naseem, I., Togneri, R. and Bennamoun, M. (2010) Linear Regression for Face Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32, 2106-2112. [Google Scholar] [CrossRef] 
[3] Wright, J., Yang, A.Y., Ganesh, A., et al. (2009) Robust Face Recognition via Sparse Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 31, 210-227. [Google Scholar] [CrossRef] 
[4] Yang, J., Zhang, L., Xu, Y. and Yang, J.Y. (2012) Beyond Sparsity: The Role of L1-Optimizer in Pattern Classification. Pattern Recognition, 45, 1104-1118. [Google Scholar] [CrossRef] 
[5] Wagner, A., Wright, J., Ganesh, A., et al. (2012) Toward a Practical Face Recognition System: Robust Registration and Illumination via Sparse Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 34, 372- 386. [Google Scholar] [CrossRef] 
[6] Zhang, L., Yang, M. and Feng, X.C. (2011) Sparse Representation or Collaborative Representation Which Helps Face Recognition? 2011 International Conference on Computer Vision, Barcelona, 6-13 November 2011, 471-478.
[7] Candés, E.J. and Tao, T. (2010) The Power of Convex Relaxation: Near-optimal Matrix Completion. IEEE Transactions on Information Theory, 56, 2053-2080. [Google Scholar] [CrossRef] 
[8] Lin, Z., Chen, M., Wu, L., et al. (2010) The Augmented Lagrange Multiplier Method for Exact Recovery of Corrupted Low-Rank Matrices. arXiv:1009.5055v2.
[9] Boyd, S., Parikh, N., Chu, E., Peleato, B. and Eckstein, J. (2011) Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers. Foundations and Trends in Machine Learning, 3, 1-122. [Google Scholar] [CrossRef] 
[10] Hansson, A., Zhang, L. and Vandenberghe, L. (2012) Subspace System Identification via Weighted Nuclear Norm Optimization. 2012 IEEE 51st IEEE Conference on Decision and Control (CDC), Maui, HI, 10-13 December 2012, 3439- 3444. [Google Scholar] [CrossRef] 
[11] Li, X., Ye, J., Hu, Y., et al. (2012) Matrix Completion by Truncated Nuclear Norm Regularization. 2012 IEEE Conference on Computer Vision and Pattern Recognition, Providence, 16-21 June 2012, 2192-2199.
[12] Lee, K.-C., Ho, J. and Kriegman, D.J. (2005) Acquiring Linear Subspaces for Face Recognition under Variable Lighting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27, 684-698. [Google Scholar] [CrossRef] 
[13] Martinez, A.M. (1998) The Ar Face Database. CVC Technical Re-port No. 24.
[14] He, R., Zheng, W.S., Hu, B.G., et al. (2011) A Regularized Correntropy Framework for Robust Pat-tern Recognition. Neural Computation, 23, 2074-2100. [Google Scholar] [CrossRef] 
[15] Yang, M., Zhang, L., Yang, J., et al. (2013) Regularized Robust Coding for Face Recognition. IEEE Transactions on Image Pro-cessing, 22, 1753-1766. [Google Scholar] [CrossRef] 
[16] He, R., Zheng, W.-S., Tan, T., et al. (2014) Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36, 261-275. [Google Scholar] [CrossRef]