Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China
Road extraction is crucial in urban planning, rescue operations, and military applications. Compared to traditional methods, using deep learning for road extraction from remote sensing images has demonstrated unique advantages. However, previous convolutional neural networks (CNN)-based road extract...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-09-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/19/4686 |
_version_ | 1797575244880281600 |
---|---|
author | Xudong Wang Yujie Cai Kang He Sheng Wang Yan Liu Yusen Dong |
author_facet | Xudong Wang Yujie Cai Kang He Sheng Wang Yan Liu Yusen Dong |
author_sort | Xudong Wang |
collection | DOAJ |
description | Road extraction is crucial in urban planning, rescue operations, and military applications. Compared to traditional methods, using deep learning for road extraction from remote sensing images has demonstrated unique advantages. However, previous convolutional neural networks (CNN)-based road extraction methods have had limited receptivity and failed to effectively capture long-distance road features. On the other hand, transformer-based methods have good global information-capturing capabilities, but face challenges in extracting road edge information. Additionally, existing excellent road extraction methods lack validation for the Chinese region. To address these issues, this paper proposes a novel road extraction model called the global–local information fusion network (GLNet). In this model, the global information extraction (GIE) module effectively integrates global contextual relationships, the local information extraction (LIE) module accurately captures road edge information, and the information fusion (IF) module combines the output features from both global and local branches to generate the final extraction results. Further, a series of experiments on two different Chinese road datasets with geographic robustness demonstrate that our model outperforms the state-of-the-art deep learning models for road extraction tasks in China. On the CHN6-CUG dataset, the overall accuracy (OA) and intersection over union (IoU) reach 97.49% and 63.27%, respectively, while on the RDCME dataset, OA and IoU reach 98.73% and 84.97%, respectively. These research results hold significant implications for road traffic, humanitarian rescue, and environmental monitoring, particularly in the context of the Chinese region. |
first_indexed | 2024-03-10T21:37:02Z |
format | Article |
id | doaj.art-9c2e1edf1e234c0b9bc58d3e1be7ee10 |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-10T21:37:02Z |
publishDate | 2023-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-9c2e1edf1e234c0b9bc58d3e1be7ee102023-11-19T14:58:33ZengMDPI AGRemote Sensing2072-42922023-09-011519468610.3390/rs15194686Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in ChinaXudong Wang0Yujie Cai1Kang He2Sheng Wang3Yan Liu4Yusen Dong5School of Computer Science, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaHubei Key Laboratory of Geological Survey and Evaluation of Ministry of Education, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaState Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan 430078, ChinaSchool of Computer Science, China University of Geosciences, Wuhan 430078, ChinaRoad extraction is crucial in urban planning, rescue operations, and military applications. Compared to traditional methods, using deep learning for road extraction from remote sensing images has demonstrated unique advantages. However, previous convolutional neural networks (CNN)-based road extraction methods have had limited receptivity and failed to effectively capture long-distance road features. On the other hand, transformer-based methods have good global information-capturing capabilities, but face challenges in extracting road edge information. Additionally, existing excellent road extraction methods lack validation for the Chinese region. To address these issues, this paper proposes a novel road extraction model called the global–local information fusion network (GLNet). In this model, the global information extraction (GIE) module effectively integrates global contextual relationships, the local information extraction (LIE) module accurately captures road edge information, and the information fusion (IF) module combines the output features from both global and local branches to generate the final extraction results. Further, a series of experiments on two different Chinese road datasets with geographic robustness demonstrate that our model outperforms the state-of-the-art deep learning models for road extraction tasks in China. On the CHN6-CUG dataset, the overall accuracy (OA) and intersection over union (IoU) reach 97.49% and 63.27%, respectively, while on the RDCME dataset, OA and IoU reach 98.73% and 84.97%, respectively. These research results hold significant implications for road traffic, humanitarian rescue, and environmental monitoring, particularly in the context of the Chinese region.https://www.mdpi.com/2072-4292/15/19/4686road extractionhigh-resolution remote sensing imagesglobal–local information fusion networkdeep learning |
spellingShingle | Xudong Wang Yujie Cai Kang He Sheng Wang Yan Liu Yusen Dong Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China Remote Sensing road extraction high-resolution remote sensing images global–local information fusion network deep learning |
title | Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China |
title_full | Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China |
title_fullStr | Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China |
title_full_unstemmed | Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China |
title_short | Global–Local Information Fusion Network for Road Extraction: Bridging the Gap in Accurate Road Segmentation in China |
title_sort | global local information fusion network for road extraction bridging the gap in accurate road segmentation in china |
topic | road extraction high-resolution remote sensing images global–local information fusion network deep learning |
url | https://www.mdpi.com/2072-4292/15/19/4686 |
work_keys_str_mv | AT xudongwang globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina AT yujiecai globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina AT kanghe globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina AT shengwang globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina AT yanliu globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina AT yusendong globallocalinformationfusionnetworkforroadextractionbridgingthegapinaccurateroadsegmentationinchina |