Low-Rank and Sparse Representation with Adaptive Neighborhood Regularization for Hyperspectral Image Classification

作     者：Zhaohui XUE Xiangyu NIE Zhaohui XUE;Xiangyu NIE

作者机构：School of Earth Sciences and EngineeringHohai UniversityNanjing 211100China 

出 版 物：《Journal of Geodesy and Geoinformation Science》 (测绘学报（英文版）)

年 卷 期：2022年第5卷第1期

页      面：73-90页

学科分类：081801[工学-矿产普查与勘探] 08[工学] 0818[工学-地质资源与地质工程] 

基　　金：National Natural Foundation of China(No.41971279) Fundamental Research Funds of the Central Universities(No.B200202012) 

主　　题：Hyperspectral Image(HSI) spectral-spatial classification Low-Rank and Sparse Representation(LRSR) Adaptive Neighborhood Regularization(ANR) 

摘      要：Low-Rank and Sparse Representation(LRSR)method has gained popularity in Hyperspectral Image(HSI)***,existing LRSR models rarely exploited spectral-spatial classification of *** this paper,we proposed a novel Low-Rank and Sparse Representation with Adaptive Neighborhood Regularization(LRSR-ANR)method for HSI *** the proposed method,we first represent the hyperspectral data via LRSR since it combines both sparsity and low-rankness to maintain global and local data structures *** LRSR is optimized by using a mixed Gauss-Seidel and Jacobian Alternating Direction Method of Multipliers(M-ADMM),which converges faster than *** to incorporate the spatial information,an ANR scheme is designed by combining Euclidean and Cosine distance metrics to reduce the mixed pixels within a ***,the predicted labels are determined by jointly considering the homogeneous pixels in the classification rule of the minimum reconstruction *** results based on three popular hyperspectral images demonstrate that the proposed method outperforms other related methods in terms of classification accuracy and generalization performance.