Spectral–Spatial Hyperspectral Image Classification Using $\ell_{1/2}$ Regularized Low-Rank Representation and Sparse Representation-Based Graph Cuts

Abstract

Hundreds of narrow contiguous spectral bands collected by a hyperspectral sensor have provided the opportunity to identify the various materials present on the surface. Moreover, spatial information, enforcing the assumption that the adjacent pixels belong to the same class with a high probability, is a valuable complement to the spectral information. In this paper, two predominant approaches have been developed to exploit the spatial information. First, by decomposing each pixel and the spatial neighborhood into a low-rank form, the spatial information can be efficiently integrated into the spectral signatures. Meanwhile, in order to describe the low-rank structure of the decomposed data more precisely, an 1/2 norm regularization is introduced and a discrete algorithm is proposed to solve the combined optimization problem by the augmented Lagrange multiplier (ALM) and a halfthreshold operator. Second, a graph cuts segmentation algorithm has been applied on the sparse-representation-based probability estimates of the hyperspectral data to further improve the spatial homogeneity of the material distribution. Experimental results on four real hyperspectral data with different spectral and spatial resolutions have demonstrated the effectiveness and versatility of the proposed spatial information-fused approaches for hyperspectral image classification.Index Terms-Augmented Lagrange multiplier (ALM), hyperspectral image classification, low-rank representation (LRR), nuclear norm.

JointL1/2-Norm Constraint and Graph-Laplacian PCA Method for Feature Extraction

“…When p ∈ [1/2,1, the smaller p is, the more effective result will be [17]. Then, Xu et al developed a simple iterative thresholding representation theory for L 1/2 -norm and obtained the desired results [18].…”

A survey on representation-based classification and detection in hyperspectral remote sensing imagery

“…The procedure ensures that the labeled samples more dissimilar to the testing pixel will provide much less contribution to its linear representation. In addition, graph-based regularization may also help with performance improvement [48,49].…”

Spectral-Spatial Hyperspectral Image Classification based on Randomized Singular Value Decomposition and 3-Dimensional Discrete Wavelet Transform

“…The first exploits the spectral and spatial information separately. In other words, the spatial dependence is extracted in advance through various spatial filters, such as morphological profiles [4][5][6], entropy [7], attribute profiles [8], and low-rank representation [9,10]. Then, these transformed spatial features are combined with the spectral features, where dimensionality reduction (DR) may be applied to perform pixel-wise classification.…”