Irregular Tensor Low-Rank Representation for Hyperspectral Image Representation

TL;DR

This paper introduces a novel irregular tensor low-rank representation model for hyperspectral images that effectively captures the inherent irregular spatial structures, improving analysis performance over existing methods.

Contribution

The paper proposes a new irregular tensor low-rank model with non-convex nuclear norm and global negative low-rank term, tailored for irregular hyperspectral image data.

Findings

Outperforms state-of-the-art methods on four public datasets

Demonstrates superior modeling of irregular tensor structures

Validates effectiveness through extensive experiments

Abstract

Spectral variations pose a common challenge in analyzing hyperspectral images (HSI). To address this, low-rank tensor representation has emerged as a robust strategy, leveraging inherent correlations within HSI data. However, the spatial distribution of ground objects in HSIs is inherently irregular, existing naturally in tensor format, with numerous class-specific regions manifesting as irregular tensors. Current low-rank representation techniques are designed for regular tensor structures and overlook this fundamental irregularity in real-world HSIs, leading to performance limitations. To tackle this issue, we propose a novel model for irregular tensor low-rank representation tailored to efficiently model irregular 3D cubes. By incorporating a non-convex nuclear norm to promote low-rankness and integrating a global negative low-rank term to enhance the discriminative ability, our proposed model is formulated as a constrained optimization problem and solved using an alternating augmented Lagrangian method. Experimental validation conducted on four public datasets demonstrates the superior performance of our method compared to existing state-of-the-art approaches. The code is publicly available at https://github.com/hb-studying/ITLRR.