An adaptive graph embedding method for feature extraction of hyperspectral images based on approximate NMR model

Abstract

This paper introduces an approximate nuclear norm based matrix regression projection (ANMRP) model, an adaptive graph embedding method, for feature extraction of hyperspectral images. The ANMRP utilizes an approximate NMR model to construct an adaptive neighborhood map between samples. The globally optimal weight matrix is obtained by optimizing the approximate NMR model using fast alternating direction method of multipliers (ADMM). The optimal projection matrix is then determined by maximizing the ratio of the local scatter matrix to the total scatter matrix, allowing for the extraction of discriminative features. Experimental results demonstrate the effectiveness of ANMRP compared to related methods.

References

[1] FUH, SUNG, RENJ, et al.. Fusion of PCA and segmented-PCA domain multiscale 2-D-SSA for effective spectral-spatial feature extraction and data classification in hyperspectral imagery[J]. IEEE transactions on geoscience and remote sensing, 2022, 60: 1-14

[2] FABIYIS D, MURRAYP, ZABALZAJ, et al.. Folded LDA: extending the linear discriminant analysis algorithm for feature extraction and data reduction in hyperspectral remote sensing[J]. IEEE journal of selected topics in applied earth observations and remote sensing, 2021, 14: 12312-12331

[3] HUANGW, XUY, HUX, et al.. Compressive hyperspectral image reconstruction based on spatial-spectral residual dense network[J]. IEEE geoscience and remote sensing, 2022, 14(17):4184

[4] HEX, YANS, HUY, et al.. Face recognition using Laplacianfaces[J]. IEEE transactions on pattern analysis and machine intelligence, 2005, 27(3): 328-340

[5] HEX, CAID, YANS, et al.. Neighborhood preserving embedding[C]. Proceedings of 2005 IEEE International Conference on Computer Vision, October 17–20, 2005, Beijing, China, 2005, New York, IEEE: 1208-12131

[6] YANGB, LIH, GUOZ. Deep manifold structure-preserving spectral-spatial feature extraction of hyperspectral image[J]. IEEE transactions on geoscience and remote sensing, 2022, 60: 1-13

[7] TASKING, CAMPS-VALLSG. Graph embedding via high dimensional model representation for hyperspectral images[J]. IEEE transactions on geoscience and remote sensing, 2022, 60: 1-11

[8] QIAOL, CHENS, TANX. Sparsity preserving projections with applications to face recognition[J]. Pattern recognition, 2010, 43(1):331-341

[9] LID, KONGF Q, WANGQ. Hyperspectral image classification via nonlocal joint kernel sparse representation based on local covariance[J]. Signal processing, 2021, 180: 1-15

[10] SHIG Y, LUOF L, TANGY M, et al.. Dimensionality reduction of hyperspectral image based on local constrained manifold structure collaborative preserving embedding[J]. Remote sensing, 2021, 13(7):1-22

[11] LYN H, DUQ, FOWLERJ E. Sparse graph-based discriminant analysis for hyperspectral imagery[J]. IEEE transactions on geoscience and remote sensing, 2014, 52(7):3872-3884

[12] YANGR C, KANJ M. Euclidean distance-based adaptive collaborative representation with Tikhonov regularization for hyperspectral image classification[J]. Multimedia tools and applications, 2023, 82(4):5823-5838

[13] YANGW, LIJ, ZHENGH, et al.. A nuclear norm based matrix regression based projections method for feature extraction[J]. IEEE access, 2018, 6: 7445-7451

[14] YANGJ, LUOL, QIANJ J, et al.. Nuclear norm based matrix regression with applications to face recognition with occlusion and illumination changes[J]. IEEE transactions on pattern analysis and machine intelligence, 2017, 39(1):156-171

[15] QIUH, WANGW L, ZHENGJ W. Sparse representation with smoothed matrix multivariate elliptical distribution[J]. Acta automatica sinica, 2019, 45(8):1548-1563

[16] QIUH, WANGR F, SUND C, et al.. A smoothed matrix multivariate elliptical distribution-based projection method for feature extraction[J]. Computational intelligence and neuroscience, 2022, 2022: 2551137

[17] HEB, YUANX. On non-ergodic convergence rate of Douglas-Rachford alternating direction method of multiplier[J]. Numerische mathematik, 2015, 130(3):567-577