Weak–Strong Graph Contrastive Learning Neural Network for Hyperspectral Image Classification

Abstract

Deep learning methods have shown promising results in various hyperspectral image (HSI) analysis tasks. Despite these advancements, existing models still struggle to accurately identify fine-classified land cover types on noisy HSIs. Traditional methods have limited performance when extracting features from noisy hyperspectral data. Graph neural networks (GNNs) offer an adaptable and robust structure by effectively extracting both spectral and spatial features. However, supervised models still require large quantities of labeled data for effective training, posing a significant challenge. Contrastive learning (CL), which leverages unlabeled data for pretraining, can mitigate this issue by reducing the dependency on extensive manual annotation. To address the issues, we propose WSGraphCL, a weak-strong graph CL model for HSI classification, and conduct experiments in a few-shot scenario. First, the image is transformed into K-hop subgraphs through a spectral-spatial adjacency matrix construction method. Second, WSGraphCL leverages CL to pretrain a graph-based encoder on the unlabeled HSI. We demonstrate that weak-strong augmentations and false negative pairs filtering stabilize pretraining and get good-quality representations. Finally, we test our model with a lightweight classifier on the features with a handful of labels. Experimental results showcase the superior performance of WSGraphCL compared to several baseline models, thereby emphasizing its efficacy in addressing the identified limitations in HSI classification. The code repository will be published on the GitHub project under the URL: https://github.com/zhu-xlab/WSGraphCL