AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau

Accurate and efficient extraction of mining areas from remote sensing imagery is essential for resource investigation, environmental assessment, and ecological management. This task is particularly crucial for the intelligent analysis of large-scale mining landscapes, such as those found on the Qing...

Full description

Saved in:
Bibliographic Details
Main Authors: Fangzhou Hong, Guojin He, Guizhou Wang, Zhaoming Zhang, Yan Peng
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/11112575/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849228524131450880
author Fangzhou Hong
Guojin He
Guizhou Wang
Zhaoming Zhang
Yan Peng
author_facet Fangzhou Hong
Guojin He
Guizhou Wang
Zhaoming Zhang
Yan Peng
author_sort Fangzhou Hong
collection DOAJ
description Accurate and efficient extraction of mining areas from remote sensing imagery is essential for resource investigation, environmental assessment, and ecological management. This task is particularly crucial for the intelligent analysis of large-scale mining landscapes, such as those found on the Qinghai–Tibet Plateau. However, existing approaches face notable limitations. Many deep learning models, especially those based on convolutional neural networks (CNNs), struggle to capture the complex and heterogeneous morphological features of mining areas in diverse geographic settings. To address these challenges, we propose AuxTransUNet, a hybrid deep learning framework that integrates CNNs with transformers to enhance both local detail extraction and global contextual understanding. The architecture incorporates a frequency-aware and boundary-guided feature fusion strategy, which improves segmentation accuracy by suppressing misclassification noise and refining boundary delineation. In addition, an auxiliary classification branch provides patch-level supervision, further strengthening semantic consistency. Extensive experiments conducted on a mining segmentation dataset covering the Qinghai–Tibet Plateau demonstrate that AuxTransUNet achieves superior performance compared to strong baseline models in terms of both segmentation accuracy and computational efficiency. The results highlight its potential as a robust and scalable solution for large-area mining monitoring using remote sensing data.
format Article
id doaj-art-91c06009be2d46f18c1909b87b3ef276
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-91c06009be2d46f18c1909b87b3ef2762025-08-22T23:09:12ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118203482035810.1109/JSTARS.2025.359559211112575AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet PlateauFangzhou Hong0https://orcid.org/0009-0003-6486-6444Guojin He1https://orcid.org/0009-0001-0128-8582Guizhou Wang2https://orcid.org/0000-0002-2347-8416Zhaoming Zhang3https://orcid.org/0000-0002-8779-2738Yan Peng4Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing, ChinaAccurate and efficient extraction of mining areas from remote sensing imagery is essential for resource investigation, environmental assessment, and ecological management. This task is particularly crucial for the intelligent analysis of large-scale mining landscapes, such as those found on the Qinghai–Tibet Plateau. However, existing approaches face notable limitations. Many deep learning models, especially those based on convolutional neural networks (CNNs), struggle to capture the complex and heterogeneous morphological features of mining areas in diverse geographic settings. To address these challenges, we propose AuxTransUNet, a hybrid deep learning framework that integrates CNNs with transformers to enhance both local detail extraction and global contextual understanding. The architecture incorporates a frequency-aware and boundary-guided feature fusion strategy, which improves segmentation accuracy by suppressing misclassification noise and refining boundary delineation. In addition, an auxiliary classification branch provides patch-level supervision, further strengthening semantic consistency. Extensive experiments conducted on a mining segmentation dataset covering the Qinghai–Tibet Plateau demonstrate that AuxTransUNet achieves superior performance compared to strong baseline models in terms of both segmentation accuracy and computational efficiency. The results highlight its potential as a robust and scalable solution for large-area mining monitoring using remote sensing data.https://ieeexplore.ieee.org/document/11112575/Convolutional neural network (CNN)–transformer hybridland use mappingopen-pit miningQinghai–Tibet plateauremote sensingsemantic segmentation
spellingShingle Fangzhou Hong
Guojin He
Guizhou Wang
Zhaoming Zhang
Yan Peng
AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Convolutional neural network (CNN)–transformer hybrid
land use mapping
open-pit mining
Qinghai–Tibet plateau
remote sensing
semantic segmentation
title AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
title_full AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
title_fullStr AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
title_full_unstemmed AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
title_short AuxTransUNet: Enhancing Remote Sensing Image Segmentation of Open-Pit Mining Areas in Qinghai–Tibet Plateau
title_sort auxtransunet enhancing remote sensing image segmentation of open pit mining areas in qinghai x2013 tibet plateau
topic Convolutional neural network (CNN)–transformer hybrid
land use mapping
open-pit mining
Qinghai–Tibet plateau
remote sensing
semantic segmentation
url https://ieeexplore.ieee.org/document/11112575/
work_keys_str_mv AT fangzhouhong auxtransunetenhancingremotesensingimagesegmentationofopenpitminingareasinqinghaix2013tibetplateau
AT guojinhe auxtransunetenhancingremotesensingimagesegmentationofopenpitminingareasinqinghaix2013tibetplateau
AT guizhouwang auxtransunetenhancingremotesensingimagesegmentationofopenpitminingareasinqinghaix2013tibetplateau
AT zhaomingzhang auxtransunetenhancingremotesensingimagesegmentationofopenpitminingareasinqinghaix2013tibetplateau
AT yanpeng auxtransunetenhancingremotesensingimagesegmentationofopenpitminingareasinqinghaix2013tibetplateau