PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing

Transformers have performed favorably in recent hyperspectral unmixing studies in which the self-attention mechanism possesses the ability to retain spectral information and spatial details. However, the lack of reliable prior information for correction guidance has resulted in an inadequate accurac...

Full description

Saved in:
Bibliographic Details
Main Authors: Yiliang Zeng, Na Meng, Jinlin Zou, Wenbin Liu
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/5/869
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850222774849634304
author Yiliang Zeng
Na Meng
Jinlin Zou
Wenbin Liu
author_facet Yiliang Zeng
Na Meng
Jinlin Zou
Wenbin Liu
author_sort Yiliang Zeng
collection DOAJ
description Transformers have performed favorably in recent hyperspectral unmixing studies in which the self-attention mechanism possesses the ability to retain spectral information and spatial details. However, the lack of reliable prior information for correction guidance has resulted in an inadequate accuracy and robustness of the network. To benefit from the advantages of the Transformer architecture and to improve the interpretability and robustness of the network, a dual-branch network with prior information correction, incorporating a Transformer network (PICT-Net), is proposed. The upper branch utilizes pre-extracted endmembers to provide pure pixel prior information. The lower branch employs a Transformer structure for feature extraction and unmixing processing. A weight-sharing strategy is employed between the two branches to facilitate information sharing. The deep integration of prior knowledge into the Transformer architecture effectively reduces endmember variability in hyperspectral unmixing and enhances the model’s generalization capability and accuracy across diverse scenarios. Experimental results from experiments conducted on four real datasets demonstrate the effectiveness and superiority of the proposed model.
format Article
id doaj-art-e616be08141940e89d56f6f8227b173a
institution OA Journals
issn 2072-4292
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-e616be08141940e89d56f6f8227b173a2025-08-20T02:06:13ZengMDPI AGRemote Sensing2072-42922025-02-0117586910.3390/rs17050869PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image UnmixingYiliang Zeng0Na Meng1Jinlin Zou2Wenbin Liu3Beijing Engineering Research Center of Industrial Spectrum Imaging, School of Automation, University of Science and Technology Beijing, Beijing 100083, ChinaBeijing Engineering Research Center of Industrial Spectrum Imaging, School of Automation, University of Science and Technology Beijing, Beijing 100083, ChinaEquipment Management and UAV Engineering School, Air Force Engineering University, Xi’an 710043, ChinaBeijing Engineering Research Center of Industrial Spectrum Imaging, School of Automation, University of Science and Technology Beijing, Beijing 100083, ChinaTransformers have performed favorably in recent hyperspectral unmixing studies in which the self-attention mechanism possesses the ability to retain spectral information and spatial details. However, the lack of reliable prior information for correction guidance has resulted in an inadequate accuracy and robustness of the network. To benefit from the advantages of the Transformer architecture and to improve the interpretability and robustness of the network, a dual-branch network with prior information correction, incorporating a Transformer network (PICT-Net), is proposed. The upper branch utilizes pre-extracted endmembers to provide pure pixel prior information. The lower branch employs a Transformer structure for feature extraction and unmixing processing. A weight-sharing strategy is employed between the two branches to facilitate information sharing. The deep integration of prior knowledge into the Transformer architecture effectively reduces endmember variability in hyperspectral unmixing and enhances the model’s generalization capability and accuracy across diverse scenarios. Experimental results from experiments conducted on four real datasets demonstrate the effectiveness and superiority of the proposed model.https://www.mdpi.com/2072-4292/17/5/869convolutional neural networkhyperspectral imageryhyperspectral unmixingspatial–spectraltransformerdual-branch network
spellingShingle Yiliang Zeng
Na Meng
Jinlin Zou
Wenbin Liu
PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
Remote Sensing
convolutional neural network
hyperspectral imagery
hyperspectral unmixing
spatial–spectral
transformer
dual-branch network
title PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
title_full PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
title_fullStr PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
title_full_unstemmed PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
title_short PICT-Net: A Transformer-Based Network with Prior Information Correction for Hyperspectral Image Unmixing
title_sort pict net a transformer based network with prior information correction for hyperspectral image unmixing
topic convolutional neural network
hyperspectral imagery
hyperspectral unmixing
spatial–spectral
transformer
dual-branch network
url https://www.mdpi.com/2072-4292/17/5/869
work_keys_str_mv AT yiliangzeng pictnetatransformerbasednetworkwithpriorinformationcorrectionforhyperspectralimageunmixing
AT nameng pictnetatransformerbasednetworkwithpriorinformationcorrectionforhyperspectralimageunmixing
AT jinlinzou pictnetatransformerbasednetworkwithpriorinformationcorrectionforhyperspectralimageunmixing
AT wenbinliu pictnetatransformerbasednetworkwithpriorinformationcorrectionforhyperspectralimageunmixing