Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images

Road extraction from UAV-based remote sensing images in rural areas presents significant challenges due to the diverse and complex characteristics of rural roads. Additionally, acquiring UAV remote sensing data for rural areas is challenging due to the high cost of equipment, the lack of clear road...

Full description

Saved in:
Bibliographic Details
Main Authors: Jian Wang, Renlong Wang, Yahui Liu, Fei Zhang, Ting Cheng
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/5/1394
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850030269755555840
author Jian Wang
Renlong Wang
Yahui Liu
Fei Zhang
Ting Cheng
author_facet Jian Wang
Renlong Wang
Yahui Liu
Fei Zhang
Ting Cheng
author_sort Jian Wang
collection DOAJ
description Road extraction from UAV-based remote sensing images in rural areas presents significant challenges due to the diverse and complex characteristics of rural roads. Additionally, acquiring UAV remote sensing data for rural areas is challenging due to the high cost of equipment, the lack of clear road boundaries requiring extensive manual annotation, and limited regional policy support for UAV operations. To address these challenges, we propose a transferable contextual network (TCNet), designed to enhance the transferability and accuracy of rural road extraction. We employ a Stable Diffusion model for data augmentation, generating diverse training samples and providing a new method for acquiring remote sensing images. TCNet integrates the clustered contextual Transformer (CCT) module, clustered cross-attention (CCA) module, and CBAM attention mechanism to ensure efficient model transferability across different geographical and climatic conditions. Moreover, we design a new loss function, the Dice-BCE-Lovasz loss (DBL loss), to accelerate convergence and improve segmentation performance in handling imbalanced data. Experimental results demonstrate that TCNet, with only 23.67 M parameters, performs excellently on the DeepGlobe and road datasets and shows outstanding transferability in zero-shot testing on rural remote sensing data. TCNet performs well on segmentation tasks without any fine-tuning for regions such as Burgundy, France, and Yunnan, China.
format Article
id doaj-art-7218e9a4214a40519c3cb990e5daafd2
institution DOAJ
issn 1424-8220
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-7218e9a4214a40519c3cb990e5daafd22025-08-20T02:59:15ZengMDPI AGSensors1424-82202025-02-01255139410.3390/s25051394Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing ImagesJian Wang0Renlong Wang1Yahui Liu2Fei Zhang3Ting Cheng4Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, ChinaFaculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, ChinaFaculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, ChinaFaculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, ChinaFaculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504, ChinaRoad extraction from UAV-based remote sensing images in rural areas presents significant challenges due to the diverse and complex characteristics of rural roads. Additionally, acquiring UAV remote sensing data for rural areas is challenging due to the high cost of equipment, the lack of clear road boundaries requiring extensive manual annotation, and limited regional policy support for UAV operations. To address these challenges, we propose a transferable contextual network (TCNet), designed to enhance the transferability and accuracy of rural road extraction. We employ a Stable Diffusion model for data augmentation, generating diverse training samples and providing a new method for acquiring remote sensing images. TCNet integrates the clustered contextual Transformer (CCT) module, clustered cross-attention (CCA) module, and CBAM attention mechanism to ensure efficient model transferability across different geographical and climatic conditions. Moreover, we design a new loss function, the Dice-BCE-Lovasz loss (DBL loss), to accelerate convergence and improve segmentation performance in handling imbalanced data. Experimental results demonstrate that TCNet, with only 23.67 M parameters, performs excellently on the DeepGlobe and road datasets and shows outstanding transferability in zero-shot testing on rural remote sensing data. TCNet performs well on segmentation tasks without any fine-tuning for regions such as Burgundy, France, and Yunnan, China.https://www.mdpi.com/1424-8220/25/5/1394remote sensingrural road extractionsemantic segmentationStable Diffusion
spellingShingle Jian Wang
Renlong Wang
Yahui Liu
Fei Zhang
Ting Cheng
Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
Sensors
remote sensing
rural road extraction
semantic segmentation
Stable Diffusion
title Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
title_full Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
title_fullStr Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
title_full_unstemmed Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
title_short Transferable Contextual Network for Rural Road Extraction from UAV-Based Remote Sensing Images
title_sort transferable contextual network for rural road extraction from uav based remote sensing images
topic remote sensing
rural road extraction
semantic segmentation
Stable Diffusion
url https://www.mdpi.com/1424-8220/25/5/1394
work_keys_str_mv AT jianwang transferablecontextualnetworkforruralroadextractionfromuavbasedremotesensingimages
AT renlongwang transferablecontextualnetworkforruralroadextractionfromuavbasedremotesensingimages
AT yahuiliu transferablecontextualnetworkforruralroadextractionfromuavbasedremotesensingimages
AT feizhang transferablecontextualnetworkforruralroadextractionfromuavbasedremotesensingimages
AT tingcheng transferablecontextualnetworkforruralroadextractionfromuavbasedremotesensingimages