Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification

The combination of convolutional neural networks and vision transformers has garnered considerable attention in hyperspectral image (HSI) classification due to their abilities to enhance the classification accuracy by concurrently extracting local and global features. However, these accuracy improve...

Full description

Saved in:
Bibliographic Details
Main Authors: Qinggang Wu, Mengkun He, Qiqiang Chen, Le Sun, Chao Ma
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10850760/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849240425486876672
author Qinggang Wu
Mengkun He
Qiqiang Chen
Le Sun
Chao Ma
author_facet Qinggang Wu
Mengkun He
Qiqiang Chen
Le Sun
Chao Ma
author_sort Qinggang Wu
collection DOAJ
description The combination of convolutional neural networks and vision transformers has garnered considerable attention in hyperspectral image (HSI) classification due to their abilities to enhance the classification accuracy by concurrently extracting local and global features. However, these accuracy improvements come at the cost of significant demands on storage resources, computational overhead, and extensive training samples. To address these challenges, this article proposes a multiscale spatial–spectral shuffling convolution integrated with a 3-D lightweight transformer (MSC-3DLT) for HSI classification. This network directly captures 3-D structural features throughout the entire feature extraction process, thereby enhancing HSI classification performance even at small sampling rates within a lightweight framework. Specifically, we first design a multiscale spatial–spectral shuffling convolution to comprehensively refine spatial–spectral feature granularities and enhance feature interactions by shuffling multiscale features across different groups. Second, to maximize the exploitation of limited training samples, we rethink transformers from the 3-D structural perspective of HSI data and propose a novel 3-D lightweight transformer (3DLT). Different from the slicing operation employed in classical transformers, the 3DLT directly extracts the inherent 3-D structural features from the HSI and mitigates quadratic complexity through a lightweight spatial–spectral pooling cross-attention mechanism. Finally, a novel training strategy is designed to adaptively adjust the learning rate based on multimetric feedback during the model training process, significantly accelerating the model fitting speed. Extensive experiments demonstrate that the proposed MSC-3DLT method remains highly competitive compared with state-of-the-art methods in terms of classification accuracy, model parameters, and floating point and operations under small sampling rates.
format Article
id doaj-art-4e752bb4463e4f8d86ce13ac0415aee1
institution Kabale University
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-4e752bb4463e4f8d86ce13ac0415aee12025-08-20T04:00:34ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01185378539410.1109/JSTARS.2025.353321110850760Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image ClassificationQinggang Wu0https://orcid.org/0009-0008-6789-8472Mengkun He1Qiqiang Chen2https://orcid.org/0009-0005-5965-7853Le Sun3https://orcid.org/0000-0001-6465-8678Chao Ma4College of Computer Science and Technology, Zhengzhou University of Light Industry, Zhengzhou, ChinaCollege of Computer Science and Technology, Zhengzhou University of Light Industry, Zhengzhou, ChinaCollege of Computer Science and Technology, Zhengzhou University of Light Industry, Zhengzhou, ChinaSchool of Computer Science and the Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing, ChinaCollege of Computer Science and Technology, Zhengzhou University of Light Industry, Zhengzhou, ChinaThe combination of convolutional neural networks and vision transformers has garnered considerable attention in hyperspectral image (HSI) classification due to their abilities to enhance the classification accuracy by concurrently extracting local and global features. However, these accuracy improvements come at the cost of significant demands on storage resources, computational overhead, and extensive training samples. To address these challenges, this article proposes a multiscale spatial–spectral shuffling convolution integrated with a 3-D lightweight transformer (MSC-3DLT) for HSI classification. This network directly captures 3-D structural features throughout the entire feature extraction process, thereby enhancing HSI classification performance even at small sampling rates within a lightweight framework. Specifically, we first design a multiscale spatial–spectral shuffling convolution to comprehensively refine spatial–spectral feature granularities and enhance feature interactions by shuffling multiscale features across different groups. Second, to maximize the exploitation of limited training samples, we rethink transformers from the 3-D structural perspective of HSI data and propose a novel 3-D lightweight transformer (3DLT). Different from the slicing operation employed in classical transformers, the 3DLT directly extracts the inherent 3-D structural features from the HSI and mitigates quadratic complexity through a lightweight spatial–spectral pooling cross-attention mechanism. Finally, a novel training strategy is designed to adaptively adjust the learning rate based on multimetric feedback during the model training process, significantly accelerating the model fitting speed. Extensive experiments demonstrate that the proposed MSC-3DLT method remains highly competitive compared with state-of-the-art methods in terms of classification accuracy, model parameters, and floating point and operations under small sampling rates.https://ieeexplore.ieee.org/document/10850760/Convolutional neural networks (CNNs)hyperspectral image (HSI) classificationmultimetric adaptive learning rate (MALR)3-D lightweight transformer (3DLT)
spellingShingle Qinggang Wu
Mengkun He
Qiqiang Chen
Le Sun
Chao Ma
Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Convolutional neural networks (CNNs)
hyperspectral image (HSI) classification
multimetric adaptive learning rate (MALR)
3-D lightweight transformer (3DLT)
title Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
title_full Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
title_fullStr Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
title_full_unstemmed Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
title_short Integrating Multiscale Spatial–Spectral Shuffling Convolution With 3-D Lightweight Transformer for Hyperspectral Image Classification
title_sort integrating multiscale spatial x2013 spectral shuffling convolution with 3 d lightweight transformer for hyperspectral image classification
topic Convolutional neural networks (CNNs)
hyperspectral image (HSI) classification
multimetric adaptive learning rate (MALR)
3-D lightweight transformer (3DLT)
url https://ieeexplore.ieee.org/document/10850760/
work_keys_str_mv AT qinggangwu integratingmultiscalespatialx2013spectralshufflingconvolutionwith3dlightweighttransformerforhyperspectralimageclassification
AT mengkunhe integratingmultiscalespatialx2013spectralshufflingconvolutionwith3dlightweighttransformerforhyperspectralimageclassification
AT qiqiangchen integratingmultiscalespatialx2013spectralshufflingconvolutionwith3dlightweighttransformerforhyperspectralimageclassification
AT lesun integratingmultiscalespatialx2013spectralshufflingconvolutionwith3dlightweighttransformerforhyperspectralimageclassification
AT chaoma integratingmultiscalespatialx2013spectralshufflingconvolutionwith3dlightweighttransformerforhyperspectralimageclassification