基于低秩子空间投影和Gabor特征的稀疏表示人脸识别算法

计算机工程与科学

基于低秩子空间投影和Gabor特征的稀疏表示人脸识别算法

杨方方1，吴锡生1，顾标准2

（1.江南大学物联网工程学院，江苏无锡 214122；2.浙江大学计算机科学与技术学院，浙江杭州 310027）

收稿日期:2015-10-08 修回日期:2015-11-26 出版日期:2017-01-25 发布日期:2017-01-25
基金资助:
国家自然科学基金(61373055)；国家科技支撑计划(2012BAH70F02)；江苏省产学研联合创新资金(BY201301535)

A face recognition algorithm based on lowrank subspace

projection and Gabor feature via sparse representation

YANG Fangfang1,WU Xisheng1,GU Biaozhun2

(1.School of Internet of Things Engineering,Jiangnan University,Wuxi 214122;

2.School of Computer Science and Technology,Zhejiang University,Hangzhou 310027,China)

Received:2015-10-08 Revised:2015-11-26 Online:2017-01-25 Published:2017-01-25

摘要/Abstract

摘要：

目前的人脸识别算法常常忽视训练过程中噪声的影响，特别是在训练数据和待测数据都受到噪声污染的情况下，识别性能会明显下降。针对含有光照变化、伪装、遮挡及表情变化等较大噪声的人脸识别问题，提出了一种基于低秩子空间投影和Gabor特征的稀疏表示人脸识别算法。该算法首先通过低秩矩阵恢复算法得到训练样本的潜在低秩结构和稀疏误差结构；然后利用主成分分析法找到低秩结构的Gabor特征所在低秩子空间的变换矩阵；再通过变换矩阵将所有样本的Gabor特征向量投影到低秩子空间上，在该低秩子空间上使用稀疏表示分类算法进行最终的分类识别。在Extend Yale B和AR数据库上的实验表明，新算法具有较高的识别率和较强的抗干扰能力。

关键词: 人脸识别, 稀疏表示, 低秩矩阵恢复, Gabor特征提取, 低秩子空间投影

Abstract:

At present, face recognition algorithms often ignore the influence of noise in the training process, especially when the training data and the testing data are corrupted, so the recognition performance may be decreased significantly. To solve the problem of face recognition with illumination variation, occlusion, camouflage and expression variation, we propose a new face recognition algorithm based on lowrank subspace projection and Gabor feature via sparse representation. Firstly, we obtain the potential low rank structure and sparse error structure of training samples by the lowrank matrix recovery algorithm, and the transformation matrix of the Gabor feature of the low rank structure by using the principal component analysis. Then the Gabor feature vectors of training samples and testing samples are projected to the low rank subspace through the transformation matrix. Finally, we use sparse representation classification algorithm to perform classification and recognition in the low rank subspace. Experiments on the Extend Yale B and AR databases show that the proposed method has a high recognition rate and strong robustness.

Key words: face recognition, sparse representation, lowrank matrix recovery, Gabor feature extraction, lowrank subspace projection

杨方方1，吴锡生1，顾标准2. 基于低秩子空间投影和Gabor特征的稀疏表示人脸识别算法[J]. 计算机工程与科学.

YANG Fangfang1,WU Xisheng1,GU Biaozhun2.

A face recognition algorithm based on lowrank subspace

projection and Gabor feature via sparse representation

[J]. Computer Engineering & Science.

参考文献［17］

［1］	Elad M,Aharon M.Image denoising via sparse and redundant representations over learned dictionaries［J］.IEEE Transactions on Image Processing,2006,15(12):37363745.
［2］	Ogawa T, Haseyama M. Missing image data reconstruction based on adaptive inverse projection via sparse representation［J］.IEEE Transactions on Multimedia,2011,13(5):974992.
［3］	Yang Jianchao,John W,Thomas H,et al.Image superresolution via sparse representation［J］.IEEE Transactions on Image Processing,2010,19(11):28612873.
［4］	Kulkarni N, Nagesh P, Gowda R, et al. Understanding compressive sensing and sparse representation based super resolution［J］.IEEE Transactions on Circuits and Systems for Video Technology,2012,22(5):778789.
［5］	Yang Bin, Li Shutao. Multifocus image fusion and restoration with sparse representation［J］.IEEE Transactions on Instrumentation and Measurement,2010,59(4):884892.
［6］	Yang Bin,Li Shutao.Pixellevel image fusion with simultaneous orthogonal matching pursuit［J］.Information Fusion,2012,13(1):1019.
［7］	John W, Yang A Y, Ganesh A, et al.Robust face recognition via sparse representation［J］.IEEE Transactions on Pattern Analysis & Machine Intelligence,2009,31(2):210227.
［8］	John W, Ma Yi, Mairal J, et al. Sparse representation for computer vision and pattern recognition［J］.Proceedings of the IEEE,2010,98(6):10311044.
［9］	OLshausen B A,Field D J.Emergence of simplecell receptive field properties by learning a sparse code for natural images［J］.Nature,1996,381(6583):607609.
［10］	Yang Meng, Zhang Lei. Gabor feature based sparse representation for face recognition with gabor occlusion dictionary［C］∥Proc of ECCV,2010:448461.
［11］	Hu Zhengping,Xu Bo,Bai Yang.Sparse representation for image recognition based on Gabor feature set and discriminative dictionary learning Gabor ［J］.Journal of Image and Graphics,2013,18(2):189194.(in Chinese)
［12］	Hu Zhengping,Li Jing.Face recognition of joint sparse representation based on lowrank subspace recovery［J］.Chinese Journal Acta Electronics Sinica,2013,41(5):987991.(in Chinese)
［13］	Candès E J, Li Xiaodong, Ma Yi, et al.Robust principal component analysis?［J］.Journal of the American Statistical Association,2009,8(1):173.
［14］	Du Haishun, Zhang Xudong, Jin Yong, et al.Face image recognition method via gabor lowrank recovery sparse representationbased classification ［J］.Acta Electronics Sinica,2014,42(12):23862393.(in Chinese)
［15］	Pati Y C,Rezaiifar R,Krishnaprasad P S.Orthogonal matching pursuit:Recursive function approximation with applications to
	wavelet decomposition［C］∥Proc of the 27th Asilomar Conference on Signals,Systems and Computers,1995:13.
［16］	Donoho D L,Tsaig Y,Ri I,et al.Sparse solution of underdetermined systems of linear equations by stagewise orthogonal matching pursuit［J］.IEEE Transactions on Information Theory,2012,58(2):10941121.
［17］	Zhang Pan,Lian Qiusheng.Lowrank relaxed collaborative representation combined with global and local features for face recognition ［J］.Journal of Computer Research and Development,2014,12(12):26632670.(in Chinese)
［18］	Lin Zhouchen,Chen Minming,Ma Yi.The augmented lagrange multiplier method for exact recovery of corrupted lowrank matrices［R］.
	Urbana, IUinois: Department of Electrical and Computer Engineering, University of IUinois at UrbanaChampaign, 2009.
［19］	Kim S J,Koh K,Lustig M,et al.An interiorpoint method for largescale l1regularized least squares ［J］.EEE Journal on Selected Topics in Signal Processing,2007,1(4):606617.
	附中文参考文献：
［11］	胡正平,徐波,白洋.Gabor特征集结合判别式字典学习的稀疏表示图像识别［J］.中国图象图形学报,2013,18(2):189194.
［12］	胡正平,李静.基于低秩子空间恢复的联合稀疏表示人脸识别算法［J］.电子学报,2013,41(5):987991.
［14］	杜海顺,张旭东,金勇,等.基于Gabor低秩恢复稀疏表示分类的人脸图像识别方法［J］.电子学报.2014,42(12):23862393.
［17］	张盼,练秋生.融合整体与局部特征的低秩松弛协作表示［J］.计算机研究与发展,2014,51(12):26632670.

[1]	严春满, 张昱瑶, 张迪. 改进标签一致KSVD字典学习的人脸识别算法[J]. 计算机工程与科学, 2022, 44(02): 291-297.
[2]	张贵仓, 王静, 苏金凤. 改进稀疏表示与积化能量和的多聚焦图像融合[J]. 计算机工程与科学, 2022, 44(01): 124-131.
[3]	何选森, 徐丽, 夏娟 . 具有相似性传播的K-均值欠定盲分离算法[J]. 计算机工程与科学, 2021, 43(10): 1880-1890.
[4]	孙雨浩, 陶洋, 胡昊. 基于判别低秩矩阵恢复和协同表示的遮挡人脸识别[J]. 计算机工程与科学, 2020, 42(12): 2199-2207.
[5]	张丽霞, 曾广平, 宣兆成. NSST域下SPCNN与SR结合的多源图像融合[J]. 计算机工程与科学, 2020, 42(11): 2042-2049.
[6]	钟国韵,王檬檬,汪宇玲,常艳荣,吴忠粱. 旋转均值跳动特征及其在模糊人脸识别中的应用[J]. 计算机工程与科学, 2020, 42(03): 474-482.
[7]	冯璋，裴东,王维. 基于改进灰狼算法优化支持向量机的人脸识别[J]. 计算机工程与科学, 2019, 41(06): 1057-1063.
[8]	李燕,章玥. 基于随机投影与加权稀疏表示残差的光照鲁棒人脸识别方法[J]. 计算机工程与科学, 2018, 40(11): 2015-2022.
[9]	纪明君，刘漫丹，才乐千. 基于半监督LDA特征子空间优化的人脸识别算法[J]. 计算机工程与科学, 2018, 40(10): 1851-1857.
[10]	王威1,2，张佳娥1,2. 基于引导滤波和shearlet稀疏的遥感图像融合算法[J]. 计算机工程与科学, 2018, 40(08): 1453-1458.
[11]	张建明1,2，廖婷婷1,2，吴宏林1,2，刘宇凯1,2. 基于改进分数阶SVD的块协作表示的小样本人脸识别算法[J]. 计算机工程与科学, 2018, 40(07): 1237-1243.
[12]	王威1,2，张佳娥1,2. 基于引导滤波和shearlet稀疏的遥感图像融合算法[J]. 计算机工程与科学, 2018, 40(07): 1250-1255.
[13]	杨培1，高雷阜1，王江1，訾玲玲2. 基于稀疏表示与字典学习的彩色图像去噪算法[J]. 计算机工程与科学, 2018, 40(05): 842-848.
[14]	杜弘彦，王士同，李滔. 基于非线性距离和夹角组合的最近特征空间嵌入方法[J]. 计算机工程与科学, 2018, 40(05): 888-897.
[15]	殷俊1,2，杨万扣3. 基于局部表示的分类方法及其人脸识别应用[J]. 计算机工程与科学, 2018, 40(03): 500-506.