Image feature extraction method based on improved nonnegative matrix factorization with universality
JIA Xu1, SUN Fuming1, LI Haojie2, CAO Yudong1
1. School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou Liaoning 121001, China; 2. School of Software Technology, Dalian University of Technology, Dalian Liaoning 116024, China
Abstract：To improve the universality of image feature extraction, an image feature extraction method based on improved Nonnegative Matrix Factorization (NMF) was proposed. Firstly, considering the practical significance of extracted image features, NMF model was used to reduce the dimension of image feature vector. Secondly, in order to represent the image by a small number of coefficients, a sparse constraint was added to the NMF model as one of the regular terms. Then, to make the optimized feature have better inter-class differentiation, the clustering property constraint would be another regular term of the NMF model. Finally, through optimizing the model by using gradient descent method, the best feature basis vector and image feature vector could be acquired. The experimental results show that for three image databases, the acquired features extracted by the improved NMF model are more conducive to correct image classification or identification, and the False Accept Rate (FAR) and False Reject Rate (FRR) are reduced to 0.021 and 0.025 respectively.
[1] SHAN H, ZHANG J, KRUGER U. Learning linear representation of sparse partitioning trees based on unsupervised kernel dimension reduction[J]. IEEE Transaction on Cybernetics, 2016, 46(12):3427-3438. [2] VLASSIS N, MOTOMURA Y, KROSE B. Supervised dimension reduction of intrinsically low-dimensional data[J]. Neural Computation, 2014, 14(1):191-215. [3] LEE D D, SEUNG H S. Learning the parts of objects by non-negative matrix factorization[J]. Nature, 1999, 401(6755):788-791. [4] HAN H, LIU S J, GAN L. Non-negativity and dependence constrained sparse coding for image classification[J]. Journal of Visual Communication and Image Representation, 2015, 26:247-254. [5] WEN J H, TIAN Z, LIU X Z, et al. Neighborhood preserving orthogonal PNMF feature extraction for hyperspectral image classification[J]. IEEE Journal of Selected Topic in Applied Earth Observations and Remote Sensing, 2013, 6(2):759-768. [6] POMPILI F, GILLIS N, ABSIL P A, et al. Two algorithms for orthogonal nonnegative matrix factorization with application to clustering[J]. Neurocomputing, 2014, 141(2):15-25. [7] KOTSIA I, ZAFEIRIOU S, PITAS I. A novel discriminant non-negative matrix factorization algorithm with applications to facial image characterization problems[J]. IEEE Transactions on Information Forensics and Security, 2007, 2(3):588-595. [8] ZDUNEK R, PHAN A H, CICHOCKI A. Image classification with nonnegative matrix factorization based on spectral projected gradient[J]. Artificial Neural Networks, 2015, 4:31-50. [9] JI Z, PANG Y, LI X. Relevance preserving projection and ranking based on one-class classification for Web image[J]. IEEE Transactions on Image Processing, 2015, 24(11):4137-4147. [10] 汪金涛,曹玉东,孙福明.稀疏约束图正则非负矩阵分解的增量学习算法[J].计算机应用,2017,37(4):1071-1074.(WANG J T, CAO Y D, SUN F M. Incremental learning algorithm based on graph regularized non-negative matrix factorization with sparseness constraints[J]. Journal of Computer Applications, 2017, 37(4):1071-1074.) [11] JIA X, SUN F M, LI H J, et al. Image multi-label annotation based on supervised nonnegative matrix factorization with new matching measurement[J]. Neurocomputing, 2017, 219(C):518-525. [12] LIU H, WU Z, CAI D, et al. Constrained nonnegative matrix factorization for image representation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2012, 34(7):1299-1311. [13] CAI D, HE X, HAN J, et al. Graph regularized nonnegative matrix factorization for data representation[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2011, 33(8):1548-1560. [14] LIU Q. Kernel local sparse representation based classifier[J]. Neural Processing Letters, 2016, 60(1):1684-1695. [15] YIN Y L, LIU L L, SUN X W. SDUMLA-HMT:a multimodal biometric database[C]//Proceedings of the 6th Chinese Conference on Biometric Recognition. Beijing:[s.n.], 2011:260-268. [16] CHUA T S, TANG J, HONG R, et al. NUSWIDE:a real-world Web image database from National University of Singapore[C]//Proceedings of the 2009 ACM International Conference on Image and Video Retrieval. New York:ACM, 2009:48. [17] 苑玮琦,王爇,孙书会.基于2DFLD的手背静脉识别算法[J].计算机应用,2010,30(3):646-649.(YUAN W Q, WANG R, SUN S H. Palm-dorsa vein recognition based on two-dimensional Fisher linear discriminant[J]. Journal of Computer Applications, 2010, 30(3):646-649.) [18] 胡学考,孙福明,李豪杰.基于稀疏约束的半监督非负矩阵分解算法[J].计算机科学,2015,42(7):280-284.(HU X K, SUN F M, LI H J. Constrained nonnegative matrix factorization with sparseness for image representation[J]. Computer Science, 2015, 42(7):280-284.)
2018, Vol. 38 
