Hyperspectral Image Spectral-Spatial Feature Extraction via Tensor Principal Component Analysis

TL;DR

This paper introduces a tensor-based spectral-spatial feature extraction method for hyperspectral images, extending PCA to its tensor form, which improves classification performance by better capturing data structure.

Contribution

The paper proposes a novel tensor PCA framework incorporating circular convolution, enhancing spectral-spatial feature extraction for hyperspectral image classification.

Findings

TPCA outperforms traditional PCA in classification accuracy

Tensor framework effectively captures spectral and spatial information

Experimental results validate the superiority of TPCA on benchmark datasets

Abstract

This paper addresses the challenge of spectral-spatial feature extraction for hyperspectral image classification by introducing a novel tensor-based framework. The proposed approach incorporates circular convolution into a tensor structure to effectively capture and integrate both spectral and spatial information. Building upon this framework, the traditional Principal Component Analysis (PCA) technique is extended to its tensor-based counterpart, referred to as Tensor Principal Component Analysis (TPCA). The proposed TPCA method leverages the inherent multi-dimensional structure of hyperspectral data, thereby enabling more effective feature representation. Experimental results on benchmark hyperspectral datasets demonstrate that classification models using TPCA features consistently outperform those using traditional PCA and other state-of-the-art techniques. These findings highlight the potential of the tensor-based framework in advancing hyperspectral image analysis.