Hyperspectral Image Classification via Dynamic Adaptive Graph Convolutional Network

Xingwen Luo

doi:10.1109/TGRS.2026.3670920

Hyperspectral Image Classification via Dynamic Adaptive Graph Convolutional Network

2026-03-05

0

PRE

AI

DOI：10.1109/TGRS.2026.3670920

Original

Original request for help

Save

Abstract

En 中文

A dynamic graph convolutional network (DGCN) can represent temporal evolutionary features. Its compatibility with the spectral-dimensional characteristics of hyperspectral images (HSIs), such as continuous gradients and local abrupt changes in spectral signatures, makes it valuable in this field. This study introduces dynamic graph modeling into HSI analysis. By mapping spectral dimensions onto virtual time series, we propose the dynamic adaptive graph convolutional network (DAGCN). The core idea is to use dynamic graph evolution to simulate continuous variations and local abrupt changes in spectral sequences, thereby capturing subtle spectral features of the Earth’s surface. The framework includes three components: the dynamic graph sequence (DGS) construction strategy, which uses superpixel segmentation and spatiotemporal graph modeling to construct slice graphs for each spectral band. The multiscale adaptive graph convolution (MAGC) module dynamically generates multiscale adjacency matrices through adaptive convolutions, learning pixel features within homogeneous regions while preserving multiscale context. The spatio-temporal graph collaborative fusion (STGC) strategy includes a dynamic segmentation-based graph update (DSG Update) module and a multispectral channel attention (MSCA) mechanism. DSG Update dynamically optimizes the graph structure of adjacent temporal phases using MAGC’s adjacency matrix, extending independent multigraph learning to spatiotemporal coupled multigraph learning for modeling spectral–spatial evolution. MSCA assigns self-attention weights to each temporal phase to enhance discriminative band selection. Experiments on public HSI datasets show that DAGCN outperforms state-of-the-art methods in classification accuracy, especially in capturing subtle spectral variations, confirming its effectiveness and superiority.

Keywords:

Attention mechanism

dynamic graph convolutional network (DGCN)

hyperspectral image (HSI) classification

superpixel segmentation

Hyperspectral Image Classification via Dynamic Adaptive Graph Convolutional Network

Abstract

Journal

Organization