Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images

The mixed pixel problem, arising from the complex vegetation types of peatlands, poses a significant challenge for remote sensing-based peatland mapping. A convolution and transformer-based reconstruction and sparse unmixing algorithm that integrates deep learning and sparse spectral unmixing is pro...

Full description

Saved in:
Bibliographic Details
Main Authors: Yulin Xu, Xiaodong Na
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/11072024/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849228490927243264
author Yulin Xu
Xiaodong Na
author_facet Yulin Xu
Xiaodong Na
author_sort Yulin Xu
collection DOAJ
description The mixed pixel problem, arising from the complex vegetation types of peatlands, poses a significant challenge for remote sensing-based peatland mapping. A convolution and transformer-based reconstruction and sparse unmixing algorithm that integrates deep learning and sparse spectral unmixing is proposed to address the spectral variability and spatial heterogeneity of the endmembers in hyperspectral datasets. Dimensionality reduction was performed on the hyperspectral data using convolution and pooling operations in a convolutional neural network. A multihead attention mechanism based on the transformer encoder captures global contextual information by correlating patches in the image, facilitating precise hyperspectral data reconstruction. Sparse spectral unmixing was applied to the reconstructed global features to obtain classification maps. Experiments were conducted on three representative peatlands in China: the Honghe, Zhalong, and Nanwenghe Nature Reserves. The proposed method outperformed the fully constrained least squares, sparsely constrained least squares spectral unmixing algorithms and deep transformer network hyperspectral unmixing, achieving a mean root-mean-square error (RMSE) of 0.1548 to 0.1781 and a mean spectral angle distance (SAD) of 0.1101 to 0.2211. The greatest improvements were 28.2% in the RMSE and 25.4% in the SAD, demonstrating the robustness and generalization capability of the algorithm.
format Article
id doaj-art-1cd4fa2916c44b05b305bb920d7b515e
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-1cd4fa2916c44b05b305bb920d7b515e2025-08-22T23:05:48ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118172971730910.1109/JSTARS.2025.358627711072024Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral ImagesYulin Xu0https://orcid.org/0009-0001-8946-6872Xiaodong Na1https://orcid.org/0000-0002-6386-4135Heilongjiang Province Key Laboratory of Geographical Environment Monitoring and Spatial Information Service in Cold Regions, Harbin Normal University, Harbin, ChinaHeilongjiang Province Key Laboratory of Geographical Environment Monitoring and Spatial Information Service in Cold Regions, Harbin Normal University, Harbin, ChinaThe mixed pixel problem, arising from the complex vegetation types of peatlands, poses a significant challenge for remote sensing-based peatland mapping. A convolution and transformer-based reconstruction and sparse unmixing algorithm that integrates deep learning and sparse spectral unmixing is proposed to address the spectral variability and spatial heterogeneity of the endmembers in hyperspectral datasets. Dimensionality reduction was performed on the hyperspectral data using convolution and pooling operations in a convolutional neural network. A multihead attention mechanism based on the transformer encoder captures global contextual information by correlating patches in the image, facilitating precise hyperspectral data reconstruction. Sparse spectral unmixing was applied to the reconstructed global features to obtain classification maps. Experiments were conducted on three representative peatlands in China: the Honghe, Zhalong, and Nanwenghe Nature Reserves. The proposed method outperformed the fully constrained least squares, sparsely constrained least squares spectral unmixing algorithms and deep transformer network hyperspectral unmixing, achieving a mean root-mean-square error (RMSE) of 0.1548 to 0.1781 and a mean spectral angle distance (SAD) of 0.1101 to 0.2211. The greatest improvements were 28.2% in the RMSE and 25.4% in the SAD, demonstrating the robustness and generalization capability of the algorithm.https://ieeexplore.ieee.org/document/11072024/Convolutional dimensionality reductionhyperspectral peatland mappingmixed pixel unmixingsparsely constrainedtransformer attention mechanism
spellingShingle Yulin Xu
Xiaodong Na
Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Convolutional dimensionality reduction
hyperspectral peatland mapping
mixed pixel unmixing
sparsely constrained
transformer attention mechanism
title Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
title_full Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
title_fullStr Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
title_full_unstemmed Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
title_short Mapping Peatlands Combing Deep Learning With Sparse Spectral Unmixing Based on Zhuhai-1 Hyperspectral Images
title_sort mapping peatlands combing deep learning with sparse spectral unmixing based on zhuhai 1 hyperspectral images
topic Convolutional dimensionality reduction
hyperspectral peatland mapping
mixed pixel unmixing
sparsely constrained
transformer attention mechanism
url https://ieeexplore.ieee.org/document/11072024/
work_keys_str_mv AT yulinxu mappingpeatlandscombingdeeplearningwithsparsespectralunmixingbasedonzhuhai1hyperspectralimages
AT xiaodongna mappingpeatlandscombingdeeplearningwithsparsespectralunmixingbasedonzhuhai1hyperspectralimages