A landslide area segmentation method based on an improved UNet

Abstract As remote sensing technology matures, landslide target segmentation has become increasingly important in disaster prevention, control, and urban construction, playing a crucial role in disaster loss assessment and post-disaster rescue. Therefore, this paper proposes an improved UNet-based l...

Full description

Saved in:
Bibliographic Details
Main Authors: Guangchen Li, Kefeng Li, Guangyuan Zhang, Ke Pan, Yuxuan Ding, Zhenfei Wang, Chen Fu, Zhenfang Zhu
Format: Article
Language:English
Published: Nature Portfolio 2025-04-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-94039-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850202644667170816
author Guangchen Li
Kefeng Li
Guangyuan Zhang
Ke Pan
Yuxuan Ding
Zhenfei Wang
Chen Fu
Zhenfang Zhu
author_facet Guangchen Li
Kefeng Li
Guangyuan Zhang
Ke Pan
Yuxuan Ding
Zhenfei Wang
Chen Fu
Zhenfang Zhu
author_sort Guangchen Li
collection DOAJ
description Abstract As remote sensing technology matures, landslide target segmentation has become increasingly important in disaster prevention, control, and urban construction, playing a crucial role in disaster loss assessment and post-disaster rescue. Therefore, this paper proposes an improved UNet-based landslide segmentation algorithm. Firstly, the feature extraction structure of the model was redesigned by integrating dilated convolution and EMA attention mechanism to enhance the model’s ability to extract image features. Additionally, this study introduces the Pag module to replace the original skip connection method, thereby enhancing information fusion between feature maps, reducing pixel information loss, and further improving the model’s overall performance. Experimental results show that compared to the original model, our model improves mIoU, Precision, Recall, and F1-score by approximately 2.4%, 2.4%, 3.2%, and 2.8%, respectively. This study not only provides an effective method for landslide segmentation tasks but also offers new perspectives for further research in related fields.
format Article
id doaj-art-9f50d407b2ef4b0a88d24d2a41dd6c94
institution OA Journals
issn 2045-2322
language English
publishDate 2025-04-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-9f50d407b2ef4b0a88d24d2a41dd6c942025-08-20T02:11:42ZengNature PortfolioScientific Reports2045-23222025-04-0115111210.1038/s41598-025-94039-5A landslide area segmentation method based on an improved UNetGuangchen Li0Kefeng Li1Guangyuan Zhang2Ke Pan3Yuxuan Ding4Zhenfei Wang5Chen Fu6Zhenfang Zhu7Shandong Jiaotong UniversityShandong Jiaotong UniversityShandong Jiaotong UniversityShandong Jiaotong UniversityShandong Jiaotong UniversityShandong Zhengyuan Yeda Environmental Technology Co., LtdShandong Jiaotong UniversityShandong Jiaotong UniversityAbstract As remote sensing technology matures, landslide target segmentation has become increasingly important in disaster prevention, control, and urban construction, playing a crucial role in disaster loss assessment and post-disaster rescue. Therefore, this paper proposes an improved UNet-based landslide segmentation algorithm. Firstly, the feature extraction structure of the model was redesigned by integrating dilated convolution and EMA attention mechanism to enhance the model’s ability to extract image features. Additionally, this study introduces the Pag module to replace the original skip connection method, thereby enhancing information fusion between feature maps, reducing pixel information loss, and further improving the model’s overall performance. Experimental results show that compared to the original model, our model improves mIoU, Precision, Recall, and F1-score by approximately 2.4%, 2.4%, 3.2%, and 2.8%, respectively. This study not only provides an effective method for landslide segmentation tasks but also offers new perspectives for further research in related fields.https://doi.org/10.1038/s41598-025-94039-5
spellingShingle Guangchen Li
Kefeng Li
Guangyuan Zhang
Ke Pan
Yuxuan Ding
Zhenfei Wang
Chen Fu
Zhenfang Zhu
A landslide area segmentation method based on an improved UNet
Scientific Reports
title A landslide area segmentation method based on an improved UNet
title_full A landslide area segmentation method based on an improved UNet
title_fullStr A landslide area segmentation method based on an improved UNet
title_full_unstemmed A landslide area segmentation method based on an improved UNet
title_short A landslide area segmentation method based on an improved UNet
title_sort landslide area segmentation method based on an improved unet
url https://doi.org/10.1038/s41598-025-94039-5
work_keys_str_mv AT guangchenli alandslideareasegmentationmethodbasedonanimprovedunet
AT kefengli alandslideareasegmentationmethodbasedonanimprovedunet
AT guangyuanzhang alandslideareasegmentationmethodbasedonanimprovedunet
AT kepan alandslideareasegmentationmethodbasedonanimprovedunet
AT yuxuanding alandslideareasegmentationmethodbasedonanimprovedunet
AT zhenfeiwang alandslideareasegmentationmethodbasedonanimprovedunet
AT chenfu alandslideareasegmentationmethodbasedonanimprovedunet
AT zhenfangzhu alandslideareasegmentationmethodbasedonanimprovedunet
AT guangchenli landslideareasegmentationmethodbasedonanimprovedunet
AT kefengli landslideareasegmentationmethodbasedonanimprovedunet
AT guangyuanzhang landslideareasegmentationmethodbasedonanimprovedunet
AT kepan landslideareasegmentationmethodbasedonanimprovedunet
AT yuxuanding landslideareasegmentationmethodbasedonanimprovedunet
AT zhenfeiwang landslideareasegmentationmethodbasedonanimprovedunet
AT chenfu landslideareasegmentationmethodbasedonanimprovedunet
AT zhenfangzhu landslideareasegmentationmethodbasedonanimprovedunet