Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jan 16, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Spatial-spectral multi-order gated aggregation network with bidirectional interactive fusion for hyperspectral image

Mingzhu Tai¹, Zhenqiu Shu¹, Songze Tang²

¹Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming, China.

Neural Networks : the Official Journal of the International Neural Network Society

|October 3, 2025

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

PrefGAN-BERT: integrating direct preference optimization into semi-supervised GAN-BERT for robust text classification.

Scientific reports·2026

Same author

Virtual Domain-Guided Cross-Modal Distillation With Multiview Correlation Awareness for Domain-Specific Multimodal Neural Machine Translation.

IEEE transactions on neural networks and learning systems·2026

Same author

scDGCL: A Dual-Level and Graph-Constrained Contrastive Learning Method for Single-Cell RNA Sequencing Data Clustering.

IEEE transactions on computational biology and bioinformatics·2026

Same author

GL-Fusion: A Multi-Omics Integration Method Based on Graph-Level Structure Fusion and Locus-Level Feature Fusion for Cancer Subtype Classification.

IEEE journal of biomedical and health informatics·2025

Same author

Spectral-Guided Multiscale Feature-Aware Transformer for Hyperspectral Image Classification.

IEEE transactions on neural networks and learning systems·2025

Same author

Disentangling Inter- and Intra-Video Relations for Multi-Event Video-Text Retrieval and Grounding.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2025

This summary is machine-generated.

This study introduces the Spatial-Spectral Multi-order Gated Aggregation Network with Bidirectional Interaction Fusion (SS-MoGAN) for hyperspectral image classification (HSIC). SS-MoGAN enhances feature extraction and classification accuracy by effectively integrating spatial and spectral information.

Area of Science:

Remote Sensing
Computer Vision
Artificial Intelligence

Background:

Convolutional Neural Networks (CNNs) show promise in hyperspectral image classification (HSIC) but struggle with efficient multi-order feature interaction and independent handling of local/global features.
Existing methods using self-attention or convolutions independently limit the complexity and efficiency of feature interaction, leading to suboptimal HSIC performance.

Purpose of the Study:

To propose a novel hyperspectral image classification (HSIC) framework, the Spatial-Spectral Multi-order Gated Aggregation Network with Bidirectional Interaction Fusion (SS-MoGAN).
To overcome the limitations of existing CNNs and self-attention mechanisms in capturing complex, multi-order spatial-spectral feature interactions for improved HSIC.

Main Methods:

Developed the Spatial-Spectral Multi-order Gated Aggregation Network with Bidirectional Interaction Fusion (SS-MoGAN), integrating convolutions and gated aggregations.

Keywords:

Bidirectional cross-attention CNNs Feature fusion Feature interaction Hyperspectral image classification Multi-order gated aggregation

More Related Videos

Author Spotlight: Unveiling the Potential of VSFG Microscopy in Studying Mesoscopically Heterogeneous Self-Assembled Structures

Author Spotlight: Unveiling the Potential of VSFG Microscopy in Studying Mesoscopically Heterogeneous Self-Assembled Structures

Published on: December 1, 2023

Related Experiment Videos

Last Updated: Jan 16, 2026

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Author Spotlight: Unveiling the Potential of VSFG Microscopy in Studying Mesoscopically Heterogeneous Self-Assembled Structures

Author Spotlight: Unveiling the Potential of VSFG Microscopy in Studying Mesoscopically Heterogeneous Self-Assembled Structures

Published on: December 1, 2023

Introduced spatial aggregation (SpaAg) and spectral aggregation (SpeAg) blocks for explicit low- and high-order feature interaction within spatial and spectral dimensions.

Incorporated bidirectional interaction fusion (BIF) blocks with cross-attention to integrate structural information and enhance fine-grained details.

Main Results:

The proposed SS-MoGAN framework demonstrated superior performance in hyperspectral image classification (HSIC) tasks.
Experiments on three benchmark datasets confirmed that SS-MoGAN outperforms existing state-of-the-art methods.
The SS-MoGAN method achieved higher accuracy in HSIC applications through efficient feature extraction and adaptive contextual processing.

Conclusions:

The SS-MoGAN framework effectively addresses limitations in current HSIC methods by enabling efficient multi-order feature extraction and adaptive contextual processing.
The integration of spatial aggregation, spectral aggregation, and bidirectional interaction fusion significantly enhances the representation of hyperspectral data for classification.
SS-MoGAN represents a significant advancement in hyperspectral image classification, offering improved accuracy and efficiency over existing approaches.