Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples

Deep learning has been widely used in landslides detection. However, in practical application, the sample quality often cannot meet the requirements of training models. Some smaller landslides are easy to be omitted if there are multiple landslide objects in one sample. Furthermore, there are some o...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaohui Liu, Ling Xu, Jinyu Zhang
Format: Article
Language:English
Published: Taylor & Francis Group 2024-12-01
Series:Geomatics, Natural Hazards & Risk
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19475705.2023.2300823
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846126441267200000
author Xiaohui Liu
Ling Xu
Jinyu Zhang
author_facet Xiaohui Liu
Ling Xu
Jinyu Zhang
author_sort Xiaohui Liu
collection DOAJ
description Deep learning has been widely used in landslides detection. However, in practical application, the sample quality often cannot meet the requirements of training models. Some smaller landslides are easy to be omitted if there are multiple landslide objects in one sample. Furthermore, there are some objects with similar shape, texture and colour to landslides (complex backgrounds), such as bare land, roads, water surfaces and artificial buildings. The traditional landslides detection method is easy to confuse landslides and complex backgrounds, which leads to false and omissive detections. To solve the above two problems, a complex background enhancement method with multi-scale samples (MSSCBE) was proposed to improve sample quality. Using the background enhanced samples, the deep learning model can not only learn differences between landslides and complex backgrounds, but also learn the multi-scale features of landslides better. The proposed method was applied to detect landslides that occurred in Jiuzhaigou County, Sichuan Province. Comparative experiments were conducted using Mask R-CNN model. And the model trained with both MSSCBE background enhanced samples and original samples has the best performance. Compared with the model trained with only original samples, Precision, Recall, F1 Score and mIoU is improved by 29.76%, 5.59%, 17.82% and 25.80%, respectively.
format Article
id doaj-art-a5eed17109a340248d49b13ef2a81e8f
institution Kabale University
issn 1947-5705
1947-5713
language English
publishDate 2024-12-01
publisher Taylor & Francis Group
record_format Article
series Geomatics, Natural Hazards & Risk
spelling doaj-art-a5eed17109a340248d49b13ef2a81e8f2024-12-12T18:11:17ZengTaylor & Francis GroupGeomatics, Natural Hazards & Risk1947-57051947-57132024-12-0115110.1080/19475705.2023.2300823Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samplesXiaohui Liu0Ling Xu1Jinyu Zhang2School of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan, Shandong, ChinaSchool of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan, Shandong, ChinaSchool of Surveying and Geo-Informatics, Shandong Jianzhu University, Jinan, Shandong, ChinaDeep learning has been widely used in landslides detection. However, in practical application, the sample quality often cannot meet the requirements of training models. Some smaller landslides are easy to be omitted if there are multiple landslide objects in one sample. Furthermore, there are some objects with similar shape, texture and colour to landslides (complex backgrounds), such as bare land, roads, water surfaces and artificial buildings. The traditional landslides detection method is easy to confuse landslides and complex backgrounds, which leads to false and omissive detections. To solve the above two problems, a complex background enhancement method with multi-scale samples (MSSCBE) was proposed to improve sample quality. Using the background enhanced samples, the deep learning model can not only learn differences between landslides and complex backgrounds, but also learn the multi-scale features of landslides better. The proposed method was applied to detect landslides that occurred in Jiuzhaigou County, Sichuan Province. Comparative experiments were conducted using Mask R-CNN model. And the model trained with both MSSCBE background enhanced samples and original samples has the best performance. Compared with the model trained with only original samples, Precision, Recall, F1 Score and mIoU is improved by 29.76%, 5.59%, 17.82% and 25.80%, respectively.https://www.tandfonline.com/doi/10.1080/19475705.2023.2300823Landslide detectioncomplex backgroundsbackground enhancementdeep learningmulti-scale samples
spellingShingle Xiaohui Liu
Ling Xu
Jinyu Zhang
Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
Geomatics, Natural Hazards & Risk
Landslide detection
complex backgrounds
background enhancement
deep learning
multi-scale samples
title Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
title_full Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
title_fullStr Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
title_full_unstemmed Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
title_short Landslide detection with Mask R-CNN using complex background enhancement based on multi-scale samples
title_sort landslide detection with mask r cnn using complex background enhancement based on multi scale samples
topic Landslide detection
complex backgrounds
background enhancement
deep learning
multi-scale samples
url https://www.tandfonline.com/doi/10.1080/19475705.2023.2300823
work_keys_str_mv AT xiaohuiliu landslidedetectionwithmaskrcnnusingcomplexbackgroundenhancementbasedonmultiscalesamples
AT lingxu landslidedetectionwithmaskrcnnusingcomplexbackgroundenhancementbasedonmultiscalesamples
AT jinyuzhang landslidedetectionwithmaskrcnnusingcomplexbackgroundenhancementbasedonmultiscalesamples