Hyperspectral Band Selection based on Generalized 3DTV and Tensor CUR Decomposition

TL;DR

This paper introduces a novel hyperspectral band selection method that combines generalized 3D total variation and tensor CUR decomposition, effectively reducing data dimensionality while preserving critical spatial-spectral information.

Contribution

The work proposes a new band selection model using G3DTV and tensor CUR decomposition, with an efficient ADMM-based algorithm, advancing spectral redundancy reduction techniques.

Findings

Outperforms existing band selection methods on benchmark datasets.

Effectively preserves spatial-spectral smoothness and intrinsic information.

Provides practical parameter selection guidelines for different noise conditions.

Abstract

Hyperspectral Imaging (HSI) serves as an important technique in remote sensing. However, high dimensionality and data volume typically pose significant computational challenges. Band selection is essential for reducing spectral redundancy in hyperspectral imagery while retaining intrinsic critical information. In this work, we propose a novel hyperspectral band selection model by decomposing the data into a low-rank and smooth component and a sparse one. In particular, we develop a generalized 3D total variation (G3DTV) by applying the $\ell_1^p$-norm to derivatives to preserve spatial-spectral smoothness. By employing the alternating direction method of multipliers (ADMM), we derive an efficient algorithm, where the tensor low-rankness is implied by the tensor CUR decomposition. We demonstrate the effectiveness of the proposed approach through comparisons with various other state-of-the-art band selection techniques using two benchmark real-world datasets. In addition, we provide practical guidelines for parameter selection in both noise-free and noisy scenarios.