Classwise Prototype-Guided Alignment Network for Cross-Scene Hyperspectral Image Classification

In the past few years, there has been significant progress in hyperspectral image classification (HSIC). However, when the trained classifier on the source scene is directly applied to a new scene, the classification performance tends to dramatically decrease because of the spectral shift phenomenon. Most existing techniques use feature alignment to learn knowledge from labeled scenes to unlabeled scenes, often overlooking the impact of noisy samples and outliers. To tackle this issue, the classwise prototype-guided alignment network (CPGAN) is proposed for cross-scene HSIC. The core idea is that classwise prototypes across scenes are employed as alignment intermediaries to guide cross-scene feature alignment. Specifically, first, spectral-spatial features from different scenes are extracted with a common feature extractor. Then, an uncertainty-aware pseudolabel selection (UPS) is designed to obtain high-confidence pseudolabels for unlabeled target scenes. Finally, a novel classwise prototype-guided alignment method is proposed to simultaneously achieve interdomain and intradomain alignment (IntraDA). The experimental results conducted on three datasets show that our method achieves superior performance compared to other cutting-edge classification algorithms.

I. Introduction

Hyperspectral imaging can capture both spatial structural information and spectral data across hundreds of contiguous bands. It has emerged as a cutting-edge technology for remote sensing community, finding widespread applications across various fields such as military reconnaissance [1], [2], [3], smart agriculture [4], [5], [6], and marine monitoring [7], [8], [9]. Hyperspectral image classification (HSIC) stands as a pivotal endeavor within hyperspectral image processing. The goal of HSIC is to determine the class of each pixel by analyzing spatial and spectral information. Over the past few decades, HSIC has garnered significant attention.