Discriminative low-rank embedding with manifold constraint for image feature extraction and classification

Abstract

The robustness against noise, outliers, and corruption is a crucial issue in image feature extraction. To address this concern, this paper proposes a discriminative low-rank embedding image feature extraction algorithm. Firstly, to enhance the discriminative power of the extracted features, a discriminative term is introduced using label information, obtaining global discriminative information and learning an optimal projection matrix for data dimensionality reduction. Secondly, manifold constraints are incorporated, unifying low-rank embedding and manifold constraints into a single framework to capture the geometric structure of local manifolds while considering both local and global information. Finally, test samples are projected into a lower-dimensional space for classification. Experimental results demonstrate that the proposed method achieves classification accuracies of 95.62%, 95.22%, 86.38%, and 86.54% on the ORL, CMUPIE, AR, and COIL20 datasets, respectively, outperforming dimensionality reduction-based image feature extraction algorithms.

References

[1] LiuZ, LuY, LaiZ, et al.. Robust sparse low-rank embedding for image dimension reduction. Applied soft computing, 2021, 113: 107907 J]

[2] MengM, LiuY, WuJ. Robust discriminant projection via joint margin and locality structure preservation. Neural processing letters, 2021, 53: 959-982 J]

[3] GeH, SongA. A new low-rank structurally incoherent algorithm for robust image feature extraction. Mathematics, 2022, 10(19):3648 J]

[4] ShahA, ChauhanY, ChaudhuryB. Principal component analysis based construction and evaluation of cryptocurrency index. Expert systems with applications, 2021, 163: 113796 J]

[5] LiH, ZhangL, HuangB, et al.. Cost-sensitive dual-bidirectional linear discriminant analysis. Information sciences, 2020, 510: 283-303 J]

[6] HeX, NiyogiP. Locality preserving projections. Advances in neural information processing systems, 2003, 16(1):186-197[J]

[7] GaoJ, LiF, WangB, et al.. Unsupervised nonlinear adaptive manifold learning for global and local information. Tsinghua science and technology, 2020, 26(2):163-171 J]

[8] FeiL, XuY, FangX, et al.. Low rank representation with adaptive distance penalty for semi-supervised subspace classification. Pattern recognition, 2017, 67: 252-262 J]

[9] GAO Y, LIN T, ZHANG Y, et al. Robust principal component analysis based on discriminant information[J]. IEEE transactions on knowledge and data engineering, 2021.

[10] XuZ, ChenC, HanG, et al.. Robust subspace clustering based on latent low rank representation with non-negative sparse Laplacian constraints. Journal of intelligent & fuzzy systems, 2021, 40(6):12151-12165 J]

[11] WongW K, LaiZ, WenJ, et al.. Low-rank embedding for robust image feature extraction. IEEE transactions on image processing, 2017, 26(6):2905-2917 J]

[12] HuW, ChenX. Low-rank projection learning based on neighbor graph. Computer engineering application, 2021, 57(7):6[J]

[13] JiangL, FangX, SunW, et al.. Low-rank constraint based dual projections learning for dimensionality reduction. Signal processing, 2023, 204: 108817 J]

[14] YangJ, YinW, ZhangY, et al.. A fast algorithm for edge-preserving variational multichannel image restoration. SIAM journal on imaging sciences, 2009, 2(2): 569-592 J]

[15] FerdinandoS S, AndyC H. Parameterisation of a stochastic model for human face identification. Proceedings of 1994 IEEE Workshop on Applications of Computer Vision, December 5–7, 1994, Sarasota, FL, USA, 1994, New York, IEEE: 138-142[C]

[16] TerenceS, SimonB, MaanB. The CMU pose, illumination, and expression (pie) database. Proceedings of 5th IEEE International Conference on Automatic Face Gesture Recognition, May 20–21, 2002, Washington DC, USA, 2002, New York, IEEE[C]

[17] MartinezA M, BenaventeR. The AR face database: CVC technical report No.24, 1998, Barcelona, Universitat Autonoma de Barcelona[R]

[18] NeneS A, NayarS K, MuraseH, et al.. Columbia object image library (coil-20), 1996, New York, Department of Computer Science, Columbia University[R]

[19] WenJ, FangX, CuiJ, et al.. Robust sparse linear discriminant analysis. IEEE transactions on circuits and systems for video technology, 2018, 29(2):390-403 J]

[20] HeX, CaiD, YanS, et al.. Neighborhood preserving embedding. 10th IEEE International Conference on Computer Vision, October 17–20, 2005, Beijing, China, 2005, New York, IEEE: 1208-1213[C]

[21] QiaoL, ChenS, TanX. Sparsity preserving projections with applications to face recognition. Pattern recognition, 2010, 43(1):331-341 J]

[22] WenJ, HanN, FangX, et al.. Low-rank preserving projection via graph regularized reconstruction. IEEE transactions on cybernetics, 2018, 49(4):1279-1291 J]