Bayesian Layer Graph Convolutioanl Network for Hyperspetral Image Classification

TL;DR

This paper introduces a Bayesian Layer Graph Convolutional Network (BLGCN) for hyperspectral image classification that enhances accuracy, generalization, and uncertainty quantification, while addressing dataset imbalance with GANs and employing a dynamic training strategy.

Contribution

The paper proposes a novel BLGCN model integrating Bayesian layers with graph convolution, combined with GAN-based data balancing and a confidence-based early stopping strategy.

Findings

Achieves high classification accuracy with strong generalization.

Effectively quantifies uncertainty in classification results.

Balances accuracy and uncertainty estimation in hyperspectral image classification.

Abstract

In recent years, research on hyperspectral image (HSI) classification has continuous progress on introducing deep network models, and recently the graph convolutional network (GCN) based models have shown impressive performance. However, these deep learning frameworks based on point estimation suffer from low generalization and inability to quantify the classification results uncertainty. On the other hand, simply applying the Bayesian Neural Network (BNN) based on distribution estimation to classify the HSI is unable to achieve high classification accuracy due to the large amount of parameters. In this paper, we design a Bayesian layer with Bayesian idea as an insertion layer into point estimation based neural networks, and propose a Bayesian Layer Graph Convolutional Network (BLGCN) model by combining graph convolution operations, which can effectively extract graph information and estimate the uncertainty of classification results. Moreover, a Generative Adversarial Network (GAN) is built to solve the sample imbalance problem of HSI dataset. Finally, we design a dynamic control training strategy based on the confidence interval of the classification results, which will terminate the training early when the confidence interval reaches the preseted threshold. The experimental results show that our model achieves a balance between high classification accuracy and strong generalization. In addition, it can quantifies the uncertainty of the classification results.