﻿ 分块稀疏表示的人脸识别研究 Face Recognition Research Based on Sparse Representation of Blocks

Software Engineering and Applications
Vol.05 No.05(2016), Article ID:18801,8 pages
10.12677/SEA.2016.55032

Face Recognition Research Based on Sparse Representation of Blocks

Yang Lu, Ruimei Kang, Fangjun Zhang

School of Computer and Information Engineering, Henan University, Kaifeng Henan

Received: Oct. 7th, 2016; accepted: Oct. 21st, 2016; published: Oct. 27th, 2016

ABSTRACT

In order to reduce the sensitivity of the face recognition algorithm to occlusion, a robust occlusion block sparse representation classification face recognition algorithm is proposed. The sparse representation algorithm uses the sparsity of high-dimensional data distribution to perform modeling, which can deal with high-dimensional image and effectively avoid dimension disaster. Block thinking is introduced in this algorithm. First of all, face image is divided into blocks which are independently sparse representation classification, and then a joint determination by all classification sub-blocks. The improved algorithm not only avoids the image feature extraction process information loss caused, but also avoids the loss of face parts information on the overall recognition results. Through simulation experiments on AR and Yale face database, it can be drawn that the improved algorithm can significantly improve the recognition rate of occluded face image, and also have some certain robustness under variable illumination.

Keywords:Sparse Representation, Face Recognition, Block, Occlusion

1. 引言

2. SRC算法

(1)

(2)

(3)

(4)

(5)

3. 分块SRC算法

① 训练和测试人脸图像按照相同的分块方式进行均匀分块；

② 将每一子块降采样为，然后将子块的所有列串联为一个列向量作为该子块的特征向量。按照这种方式处理所有训练图像，则每一子块都可以形成一个过完备词典矩阵Ab

③ 第一个子块对应的过完备词典建立后，测试图片Y的第一个子块降采样得到该子块的特征向量，然后求解Yb的稀疏表达式，比较得到的子块残差rk对子块进行判别分类，并计算最小残差r1和次最小残差r2的比值。对Y的剩余子块重复该步骤；

Figure 1. The illustration based on sparse representation of blocks

④ 将所有子块所得类别结果和所求残差比值结果分别以向量形式保存；

⑤ 统计测试图片Y中所有子块所属类别，按照“少数服从多数”准则进行判断，即类别数量最多的为Y的识别结果；若有两个或两个以上类别数量相同或者所有子块均属于不同类时，则将具有最大残差比值的那一子块所属类别确定为测试图片所属类别。

4. 实验结果与分析

4.1. Extended Yale B人脸库实验结果

4.2. AR人脸库实验结果

Figure 2. Face images with different occlusion area

Figure 3. SRC and improved algorithm recognition rate comparison on different occlusion area

Figure 4. The recognition rate comparison on different Resolution and block mode

Figure 5. The recognition time comparison on different Resolution and block mode

Table 1. The recognition rate comparison about different algorithm

4.3. 实验结果分析

5. 结束语

Face Recognition Research Based on Sparse Representation of Blocks[J]. 软件工程与应用, 2016, 05(05): 277-284. http://dx.doi.org/10.12677/SEA.2016.55032

1. 1. Hua, G., Yang, M.H., Learnedmiller, E., et al. (2011) Introduction to the Special Section, on Real-World Face Recognition. IEEE Transactions on Pattern Analysis & Machine Intelligence, 33, 1921-1924. http://dx.doi.org/10.1109/TPAMI.2011.182

2. 2. 林玲. 基于部分遮挡人脸识别算法的研究[J]. 计算机仿真, 2012, 29(1): 231-233.

3. 3. Wu, J. and Trivedi, M.M. (2006) A Regression Model in Tensor PCA Subspace for Face Image Super-Resolution Reconstruction. 18th International Conference on Pattern Recognition, Hong Kong, 20-24 August 2006, 627-630.

4. 4. 王志明, 陶建华. 人脸遮挡区域检测与重建[J]. 計算機研究與發展,2010,47(1):16-22.

5. 5. Vo, N., Vo, D., Challa, S., et al. (2009) Compressed Sensing for Face Recognition. IEEE Symposium on Computational Intelligence for Image Processing, Nashville, 30 March-2 April 2009, 104-109. http://dx.doi.org/10.1109/ciip.2009.4937888

6. 6. Wagner, A., Wright, J., Ganesh, A., et al. (2012) Toward a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representation. IEEE Transactions on Pattern Analysis & Machine Intelligence, 34, 372-386. http://dx.doi.org/10.1109/TPAMI.2011.112

7. 7. Wright, J., Yang, A.Y., Ganesh, A., et al. (2008) Robust Face Recognition via Sparse Representation. IEEE Transactions on Pattern Analysis & Machine Intelligence, 31, 2368-2378. http://dx.doi.org/10.1109/afgr.2008.4813404

8. 8. 朱明旱, 李树涛, 叶华. 稀疏表示分类中遮挡字典构造方法的改进[J]. 计算机辅助设计与图形学学报, 2014(11): 2064-2069.

9. 9. Yang, J., Yu, K., Gong, Y., et al. (2009) Linear Spatial Pyramid Matching Using Sparse Coding for Image Classification. Computer Vision and Pattern Recognition, Miami, 1794-1801.

10. 10. Candè, E.J. and Wakin, M.B. (2008) An Introduction to Compressive Sampling. IEEE Signal Processing Magazine, 25, 21-30. http://dx.doi.org/10.1109/MSP.2007.914731

11. 11. Yang, A.Y., Zhou, Z., Balasubramanian, A.G., et al. (2013) Fast-Minimization Algorithms for Robust Face Recognition. IEEE Transactions on Image Processing, 22, 3234-3246. http://dx.doi.org/10.1109/TIP.2013.2262292

12. 12. Wang, X.Z., Yan, L.I., Guo, L.H., et al. (2009) Face Recognition Algorithm Based on BD-PCA and K-NN. Journal of Wuhan University of Technology, 40-41, 130-133.

13. 13. Liang, S., Wang, Y. and Liu, Y. (2012) Face Recognition Algorithm Based on Compressive Sensing and SRC. Second International Conference on Instrumentation, Measurement, Computer, Communication and Control, 1460-1463. http://dx.doi.org/10.1109/IMCCC.2012.342

14. 14. Deng, W., Hu, J. and Guo, J. (2013) In Defense of Sparsity Based Face Recognition. IEEE Conference on Computer Vision and Pattern Recognition. 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Portland, 23-28 June 2013, 399-406. http://dx.doi.org/10.1109/CVPR.2013.58