基于核稀疏表示的多模身份识别算法
|
摘要
针对训练样本与测试样本非线性可分问题,借助核算法,将样本特征向量映射到易实现线性可分的核空间,进而在高维核空间内运用核稀疏表示对所提取的特征进行分类表达。该算法受益于将核稀疏表示理论同多模生物识别技术相结合,使其对生物特征图像有较强的鲁棒性。实验证明基于核稀疏表示的多模身份识别算法在遮挡、含噪声的情况下具有较好的识别准确率,相较于其他同类算法在性能上有一定程度的提高。
Aiming at the problem of non-linear separability between training sample and test sample,this paper uses the kernel arithmetic to map the sample eigenvector to the kernel space which is easy to realize linear separability,and then the kernel sparse representation is used in the high-dimensional kernel space to classify the extracted features. This algorithm benefits from the combination of kernel sparse representation theory with multi-modal biometrics,which results in robustness to biometric images. Experiments show that the multi-modal identification algorithm based on kernel sparse representation has better recognition accuracy under occlusion and noises,which improves the performance to some extent compared with other similar algorithms.
Aiming at the problem of non-linear separability between training sample and test sample,this paper uses the kernel arithmetic to map the sample eigenvector to the kernel space which is easy to realize linear separability,and then the kernel sparse representation is used in the high-dimensional kernel space to classify the extracted features. This algorithm benefits from the combination of kernel sparse representation theory with multi-modal biometrics,which results in robustness to biometric images. Experiments show that the multi-modal identification algorithm based on kernel sparse representation has better recognition accuracy under occlusion and noises,which improves the performance to some extent compared with other similar algorithms.
引文
[1] Di Wei,Zhang Lei,Zhang D,et al.Studies on Hy-perspectral Face Recognition in Visible SpectrumwithFeatureBandSelection.IEEETransonSystems,Man and Cybernetics,2010,40(6):1354-1361.
[2] Yang Meng,Zhang Lei,Yang Jian,et al.MetafaceLearning for Sparse Representation Based FaceRecognition//Proc of the 17th IEEE InternationalConference on Image Processing. Hong Kong,Chi-na,2010:1601-1604.
[3]杨方方,吴锡生,顾标准.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(1):131-137.
[4]马晓,张番栋,封举富.基于深度学习特征的稀疏表示的人脸识别方法[J].智能系统学报,2016,11(3):279-286.
[5] Deng W H,Hu J,Guo J. Extended SRC:UnderSampled Face Recognition via Intra-class VariantDictionary[J].IEEE Transactions on Pattern Analy-sis and Machine Intelligence,2012,34(9):1864-1870.
[6] Elhami E,Vida1 R. Block-sparse Recovery viaConvex Optimization[J]. IEEE Transactions on Sig-nal Processing,2012,60(8):4049-4107.
[7] Zhang L,Zhou W,Chang P.Kernel Sparse Repre-sentation-based Classifier[J]. IEEE Transactionson Signal Processing,2012,60(4):1684-1694.
[8] Gao S,Tsang I,Chia L T.Sparse Representationwith Kernels[J]. IEEE Transactions on Image Pro-cessing,2013,22(2):423-434.
[9]廖瑞华,李勇帆,刘宏.基于稳健主成分分析与核稀疏表示的人脸识别[J].计算机工程,2016,42(2):200-205.
[10]Sayan Banerjee,Amitava Chatterjee. Robust multi-modal multivariate ear recognition using kernelbased simultaneous sparse representation[J]. Engi-neering Applications of Artificial Intelligence,2017:64.
[11]Gregory Ditzler,Nidhal Carla Bouaynaya,RomanShterenberg. AKRON:An algorithm for approximat-ing sparse kernel reconstruction[J]. Signal Process-ing,2018:144.
[12]黄不了,刘明明,孙伟,等.基于谱回归的核稀疏表示分类方法[J].计算机应用,2017,37(S1):97-102.
[13]王玉伟,董西伟,陈芸.基于稀疏表示的多模态生物特征识别算法[J].计算机工程,2016,42(10):219-225.
[14]A.Tekade, S.P.Narote.Feature Fusion methodbased on Fisher Discriminant Analysis for Faceand Ear for Multimodal Recognition[J].Internation-al Journal of Engineering Research&Technology,2012:2278-0181.
[15]X.Xu,Z.Mu,Feature fusion method based on KC-CA for ear and profile face based multimodal recog-nition[C]. Proceedings of the IEEE InternationalConference on Automation and Logistics,August2007:620-623.
[16]刘付民,张治斌,沈记全.核典型相关分析算法的多特征融合情感识别[J].计算机工程与应用,2014,50(9):193-196,253.
[17]李波.基于稀疏表示的人脸人耳多模身份识别研究[D].青岛:中国石油大学(华东),2016.
[2] Yang Meng,Zhang Lei,Yang Jian,et al.MetafaceLearning for Sparse Representation Based FaceRecognition//Proc of the 17th IEEE InternationalConference on Image Processing. Hong Kong,Chi-na,2010:1601-1604.
[3]杨方方,吴锡生,顾标准.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(1):131-137.
[4]马晓,张番栋,封举富.基于深度学习特征的稀疏表示的人脸识别方法[J].智能系统学报,2016,11(3):279-286.
[5] Deng W H,Hu J,Guo J. Extended SRC:UnderSampled Face Recognition via Intra-class VariantDictionary[J].IEEE Transactions on Pattern Analy-sis and Machine Intelligence,2012,34(9):1864-1870.
[6] Elhami E,Vida1 R. Block-sparse Recovery viaConvex Optimization[J]. IEEE Transactions on Sig-nal Processing,2012,60(8):4049-4107.
[7] Zhang L,Zhou W,Chang P.Kernel Sparse Repre-sentation-based Classifier[J]. IEEE Transactionson Signal Processing,2012,60(4):1684-1694.
[8] Gao S,Tsang I,Chia L T.Sparse Representationwith Kernels[J]. IEEE Transactions on Image Pro-cessing,2013,22(2):423-434.
[9]廖瑞华,李勇帆,刘宏.基于稳健主成分分析与核稀疏表示的人脸识别[J].计算机工程,2016,42(2):200-205.
[10]Sayan Banerjee,Amitava Chatterjee. Robust multi-modal multivariate ear recognition using kernelbased simultaneous sparse representation[J]. Engi-neering Applications of Artificial Intelligence,2017:64.
[11]Gregory Ditzler,Nidhal Carla Bouaynaya,RomanShterenberg. AKRON:An algorithm for approximat-ing sparse kernel reconstruction[J]. Signal Process-ing,2018:144.
[12]黄不了,刘明明,孙伟,等.基于谱回归的核稀疏表示分类方法[J].计算机应用,2017,37(S1):97-102.
[13]王玉伟,董西伟,陈芸.基于稀疏表示的多模态生物特征识别算法[J].计算机工程,2016,42(10):219-225.
[14]A.Tekade, S.P.Narote.Feature Fusion methodbased on Fisher Discriminant Analysis for Faceand Ear for Multimodal Recognition[J].Internation-al Journal of Engineering Research&Technology,2012:2278-0181.
[15]X.Xu,Z.Mu,Feature fusion method based on KC-CA for ear and profile face based multimodal recog-nition[C]. Proceedings of the IEEE InternationalConference on Automation and Logistics,August2007:620-623.
[16]刘付民,张治斌,沈记全.核典型相关分析算法的多特征融合情感识别[J].计算机工程与应用,2014,50(9):193-196,253.
[17]李波.基于稀疏表示的人脸人耳多模身份识别研究[D].青岛:中国石油大学(华东),2016.