基于QR分解重构虚拟样本的人脸识别算法

计算机工程与科学

基于QR分解重构虚拟样本的人脸识别算法

郭艳君，许道云，秦永彬

（贵州大学计算机科学与技术学院，贵州贵阳 550025）

收稿日期:2016-07-10 修回日期:2016-09-15 出版日期:2016-11-25 发布日期:2016-11-25
基金资助:
国家自然科学基金（61262006，61540050）；贵州省重大应用基础研究项目（黔科合JZ字［2014］2001号）；贵州省科技厅联合基金（黔科合LH字［2014］7636号）

A face recognition algorithm based on QR

decomposition and reconstruction of virtual samples

GUO Yanjun，XU Daoyun，QIN Yongbin

（School of Computer Science and Technology,Guizhou University,Guiyang 550025,China）

Received:2016-07-10 Revised:2016-09-15 Online:2016-11-25 Published:2016-11-25

摘要/Abstract

摘要：

一直以来，小样本问题是人脸识别应用面临的一大难题。针对在实际人脸识别过程中存在的样本不足的问题，首次提出基于QR分解重构虚拟训练样本的算法。该算法使用Q与R的部分信息构造出与原始人脸图像具有一定差异性的虚拟样本，增加了人脸图像更多可能性变化的有效特征，扩大了训练样本集，然后对原始样本和虚拟重构样本协同表示的结果进行加权融合，选取最优权重组合，调整原始样本与虚拟样本对结果的影响比重，得到正确识别率。以ORL、FERET和AR三大人脸数据库对算法进行实验验证。实验结果表明，此算法能够取得较高的识别准确率。

关键词: 人脸识别, QR分解, 虚拟样本, 协同表示

Abstract:

The small sample problem has been a major challenge for face recognition applications for a long time. Concerning the insufficient sample problem in the face recognition process, we propose a face recognition algorithm based on QR decomposition and reconstruction of virtual training samples. We use partial Q and R information to construct virtual samples which have certain difference from the original face image, thus more effective characteristics of the face image that likely change with emotions and illumination are increased, and the training sample sets are expanded. Then we perform weighting fusion on the collaborative representations of the original samples and virtual samples, select the optimal weight combination, adjust the proportions of the original sample and virtual sample which can impact on the results, and obtain the correct recognition rate. Experiments on the ORL, FERET and AR face data sets show that the proposed algorithm can achieve a higher recognition accuracy.

Key words: face recognition, QR decomposition, virtual samples；collaborative representation

郭艳君，许道云，秦永彬. 基于QR分解重构虚拟样本的人脸识别算法[J]. 计算机工程与科学.

GUO Yanjun，XU Daoyun，QIN Yongbin.

A face recognition algorithm based on QR

decomposition and reconstruction of virtual samples

[J]. Computer Engineering & Science.

参考文献

［1］Naseem I, Togneri R, Bennamoun M. Linear regression for face recognition［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32（11）:21062112.

［2］Wright J,Yang A Y,Ganesh A,et al.Robust face recognition via sparse representation［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31（2）:210227．

［3］Zhang L, Yang M, Feng X, et al. Collaborative representation based classification for face recognition［J］. 2012:arXiv:12042358.

［4］Zhang L,Yang M,Feng X.Sparse representation or collaborative representation:Which helps face recognition？［C］∥Proc of 2011 IEEE International Conference on Computer Vision （ICCV）,2011:471478．

［5］Xu Y,Zhu X,Li Z,et al.Using the original and ‘symmetrical face’ training samples to perform representation based twostep face recognition［J］.Pattern Recognition,2013,46（4）:11511158．

［6］Xu Y,Li X,Yang J,et al.Integrating conventional and inverse representation for face recognition［J］.IEEE Transactions on Cybernetics,2014,44（10）:17381746．

［7］Xu Y,Li X,Yang J,et al.Integrate the original face image and its mirror image for face recognition［J］.Neurocomputing,2014,131:191199．

［8］Ryu Y S,Oh S Y.Simple hybrid classifier for face recognition with adaptively generated virtual data［J］.Pattern Recognition Letters,2002,23（7）:833841．

［9］Sharma A,Dubey A,Tripathi P,et al.Pose invariant virtual classifiers from single training image using novel hybrideigenfaces［J］.Neurocomputing,2010,73（10）:18681880．

［10］Beymer D， Poggio T. Face recognition from one example view［C］∥Proc of the 5th IEEE International Conference on Computer Vision,1995:500507.

［11］Vetter T.Synthesis of novel views from a single face image［J］.International Journal of Computer Vision,1998,28（2）:103116．

［12］Bin T, Siwei L, Hua H.High performance face recognition system by creating virtual sample［C］∥Proc of the 2003 IEEE International Conference on Neural Networks and Signal Processing,2003:972975．

［13］Jung H C,Hwang B W,Lee S W.Authenticating corrupted face image based on noise model［C］∥Proc of the 6th IEEE International Conference on Automatic Face and Gesture Recognition,2004:272277．

［14］Thian N P H,Marcel S,Bengio S.Improving face authentication using virtual samples［C］∥Proc of 2003 IEEE International Conference on Acoustics,Speech,and Signal Processing（ICASSP’03）,2003:III2336．

［15］Poggio T,Vetter T.Recognition and structure from one 2D model view:Observations on prototypes,object classes and symmetries［R］. MA:Massachusetts Institute of Technology, Cambridge, 1992．

［16］Wen Jinwei, Luo Siwei, Zhao Jiali, et al. A Small sample face recognition statistical learning method based on virtual samples［J］.Journal of Computer Research and Development, 2002,39(7):814818.(in Chinese)

［17］Wang Xiaodong, Guo Lei, Fang Jun. Research on ontologydriven text virtual sample constructing［J］.Computer Science, 2008,35(3):142145.(in Chinese)

［18］Yu Xu, Yang Jing, Xie Zhiqiang. Research on virtual sample generation technology ［J］.Computer Science,2011,38(3):1619.(in Chinese)

［19］Donoho D L.Compressed sensing［J］.IEEE Transactions on Information Theory,2006,52（4）:12891306．

［20］Zheng Xiao, Bo Hua, Sun Qiang. An algorithm for measurement matrix based on QR decomposition and eigenvalue optimization ［J］.CAAI Transactions on Intelligent Systems, 2015,10(1):149155.(in Chinese)

附中文参考文献：

［16］温津伟，罗四维，赵嘉莉，等.通过创建虚拟样本的小样本人脸识别统计学习方法［J］.计算机研究与发展,2002，39（7）:814818.

［17］王晓东，郭雷，方俊.本体驱动的文本虚拟样本构造方法研究［J］.计算机科学，2008，35（3）:142145．

［18］于旭，杨静，谢志强.虚拟样本生成技术研究［J］.计算机科学，2011，38（3）:1619．

［20］郑晓，薄华，孙强.QR分解与特征值优化观测矩阵的算法研究［J］.智能系统学报，2015，10（1）:149155．

[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]	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.
[3]	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.
[4]	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.
[5]	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.
[6]	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.
[7]	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.
[8]	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.
[9]	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.
[10]	刘冀伟1，史尹嘉1，白羽1，严朝雯2. 一种脑电信息监督人脸图像的注意力分析方法[J]. 计算机工程与科学, 2018, 40(02): 298-303.
[11]	刘欢，苏士美. 基于EHMM-SVM的人脸识别算法研究[J]. 计算机工程与科学, 2017, 39(05): 925-930.
[12]	崔鹏，张雪婷. 一种基于块共同特征值的人脸识别方法[J]. 计算机工程与科学, 2017, 39(04): 777-784.
[13]	丁铭，贾维敏. 基于L1范数的分块二维局部保持投影算法[J]. 计算机工程与科学, 2017, 39(03): 519-523.
[14]	傅俊鹏，陈秀宏，葛骁倩. L2,1范数正则化的不相关判别分析及其在人脸识别中的应用[J]. 计算机工程与科学, 2017, 39(02): 343-350.
[15]	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, 2017, 39(01): 131-137.