Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification

Abstract Efficient extraction of spectral-spatial features is essential for accurate hyperspectral image (HSI) classification, where capturing both local texture and global semantic relationships is critical. While Convolutional Neural Networks (CNNs) and Transformers have shown strong capabilities...

Full description

Saved in:
Bibliographic Details
Main Authors: Tahir Arshad, Bo peng, Ali Rahman, Rahim khan, Sajid Ullah khan, Sultan Alnazi, Nazik Alturki
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-12489-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849763773144891392
author Tahir Arshad
Bo peng
Ali Rahman
Rahim khan
Sajid Ullah khan
Sultan Alnazi
Nazik Alturki
author_facet Tahir Arshad
Bo peng
Ali Rahman
Rahim khan
Sajid Ullah khan
Sultan Alnazi
Nazik Alturki
author_sort Tahir Arshad
collection DOAJ
description Abstract Efficient extraction of spectral-spatial features is essential for accurate hyperspectral image (HSI) classification, where capturing both local texture and global semantic relationships is critical. While Convolutional Neural Networks (CNNs) and Transformers have shown strong capabilities in modeling local and global dependencies, most existing architectures operate directly on raw spectral-spatial inputs and lack explicit mechanisms for frequency-domain decomposition thereby overlooking potentially discriminative phase and frequency components. To address this limitation, we propose a Spectral-Spatial Wave and Frequency Interactive Transformer for HSI classification, which integrates frequency-aware and phase-aware token representations into a unified Transformer framework. Specifically, our model first employs a CNN backbone to extract shallow spectral-spatial features. These are then processed by a novel Frequency Domain Transformer Encoder, composed of two complementary branches: (i) a Spectral-Spatial Frequency Generator that extracts multiscale frequency features, and (ii) a Spectral-Spatial Wave Generator that encodes phase and amplitude characteristics as complex-valued wave tokens. A Spectral-Spatial Interaction Module fuses these components, followed by a Local-Global Modulator that refines semantic representations from multiple perspectives. Extensive experiments on five benchmark HSI datasets, demonstrate the effectiveness of our approach. The proposed model achieves state-of-the-art classification performance, with Overall Accuracies of 98.49%, 98.60%, 99.07%, 98.29%, and 97.97%, consistently outperforming existing methods.
format Article
id doaj-art-e564d5edb237416cb037f36176b2c281
institution DOAJ
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e564d5edb237416cb037f36176b2c2812025-08-20T03:05:18ZengNature PortfolioScientific Reports2045-23222025-07-0115112010.1038/s41598-025-12489-3Spectral-spatial wave and frequency interactive transformer for hyperspectral image classificationTahir Arshad0Bo peng1Ali Rahman2Rahim khan3Sajid Ullah khan4Sultan Alnazi5Nazik Alturki6School of Computing and Artificial Intelligence, Southwest Jiaotong UniversitySchool of Computing and Artificial Intelligence, Southwest Jiaotong UniversitySchool of Civil Engineering, Faculty of Engineering and Physical Sciences, University of LeedsCollege of Information and Communication Engineering, Harbin Engineering UniversityDepartment of Information Systems, College of Computer Engineering and Sciences, Prince Sattam bin Abdulaziz UniversityDepartment of Information Systems, College of Computer Engineering and Sciences, Prince Sattam bin Abdulaziz UniversityDepartment of Information Systems, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman UniversityAbstract Efficient extraction of spectral-spatial features is essential for accurate hyperspectral image (HSI) classification, where capturing both local texture and global semantic relationships is critical. While Convolutional Neural Networks (CNNs) and Transformers have shown strong capabilities in modeling local and global dependencies, most existing architectures operate directly on raw spectral-spatial inputs and lack explicit mechanisms for frequency-domain decomposition thereby overlooking potentially discriminative phase and frequency components. To address this limitation, we propose a Spectral-Spatial Wave and Frequency Interactive Transformer for HSI classification, which integrates frequency-aware and phase-aware token representations into a unified Transformer framework. Specifically, our model first employs a CNN backbone to extract shallow spectral-spatial features. These are then processed by a novel Frequency Domain Transformer Encoder, composed of two complementary branches: (i) a Spectral-Spatial Frequency Generator that extracts multiscale frequency features, and (ii) a Spectral-Spatial Wave Generator that encodes phase and amplitude characteristics as complex-valued wave tokens. A Spectral-Spatial Interaction Module fuses these components, followed by a Local-Global Modulator that refines semantic representations from multiple perspectives. Extensive experiments on five benchmark HSI datasets, demonstrate the effectiveness of our approach. The proposed model achieves state-of-the-art classification performance, with Overall Accuracies of 98.49%, 98.60%, 99.07%, 98.29%, and 97.97%, consistently outperforming existing methods.https://doi.org/10.1038/s41598-025-12489-3Attention moduleConvolutional neural networkHyperspectral image classificationFrequency domainVision transformer
spellingShingle Tahir Arshad
Bo peng
Ali Rahman
Rahim khan
Sajid Ullah khan
Sultan Alnazi
Nazik Alturki
Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
Scientific Reports
Attention module
Convolutional neural network
Hyperspectral image classification
Frequency domain
Vision transformer
title Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
title_full Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
title_fullStr Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
title_full_unstemmed Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
title_short Spectral-spatial wave and frequency interactive transformer for hyperspectral image classification
title_sort spectral spatial wave and frequency interactive transformer for hyperspectral image classification
topic Attention module
Convolutional neural network
Hyperspectral image classification
Frequency domain
Vision transformer
url https://doi.org/10.1038/s41598-025-12489-3
work_keys_str_mv AT tahirarshad spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT bopeng spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT alirahman spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT rahimkhan spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT sajidullahkhan spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT sultanalnazi spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification
AT nazikalturki spectralspatialwaveandfrequencyinteractivetransformerforhyperspectralimageclassification