The field of hyperspectral image analysis and graph neural networks is rapidly evolving, with a focus on developing more accurate and efficient methods for image classification, object detection, and graph representation learning. Recent research has highlighted the importance of incorporating spectral and spatial information into deep learning models, as well as the need for more robust and generalizable methods for handling complex and high-dimensional data. Notable papers in this area include one that proposes a novel open-set domain generalization framework for hyperspectral image classification, which combines spectral-invariant frequency disentanglement, dual-channel residual networks, evidential deep learning, and spectral-spatial uncertainty disentanglement to achieve state-of-the-art performance. Another noteworthy paper presents a comprehensive study of Transformer-based models for hyperspectral image classification, providing a thorough review of existing methods and identifying key challenges and open problems in the field.
Advances in Hyperspectral Image Analysis and Graph Neural Networks
Sources
Spectral Graph Neural Networks are Incomplete on Graphs with a Simple Spectrum
SDS-Net: Shallow-Deep Synergism-detection Network for infrared small target detection
A Novel Large-scale Crop Dataset and Dual-stream Transformer Method for Fine-grained Hierarchical Crop Classification from Integrated Hyperspectral EnMAP Data and Multispectral Sentinel-2 Time Series
Hyperspectral Image Classification via Transformer-based Spectral-Spatial Attention Decoupling and Adaptive Gating
Transformers Meet Hyperspectral Imaging: A Comprehensive Study of Models, Challenges and Open Problems
Evidential Deep Learning with Spectral-Spatial Uncertainty Disentanglement for Open-Set Hyperspectral Domain Generalization
Structural-Spectral Graph Convolution with Evidential Edge Learning for Hyperspectral Image Clustering
California Crop Yield Benchmark: Combining Satellite Image, Climate, Evapotranspiration, and Soil Data Layers for County-Level Yield Forecasting of Over 70 Crops
FSATFusion: Frequency-Spatial Attention Transformer for Infrared and Visible Image Fusion
It's Not the Target, It's the Background: Rethinking Infrared Small Target Detection via Deep Patch-Free Low-Rank Representations
Boosting Adversarial Transferability for Hyperspectral Image Classification Using 3D Structure-invariant Transformation and Intermediate Feature Distance
Class-Incremental Learning for Honey Botanical Origin Classification with Hyperspectral Images: A Study with Continual Backpropagation