A New Mask R-CNN-Based Method for Improved Landslide Detection
This article presents a novel method of landslide detection by exploiting the Mask R-CNN capability of identifying an object layout by using a pixel-based segmentation, along with transfer learning used to train the proposed model. A data set of 160 elements is created containing landslide and nonla...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9373966/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849223123705004032 |
|---|---|
| author | Silvia Ullo Amrita Mohan Alessandro Sebastianelli Shaik Ahamed Basant Kumar Ramji Dwivedi Ganesh R. Sinha |
| author_facet | Silvia Ullo Amrita Mohan Alessandro Sebastianelli Shaik Ahamed Basant Kumar Ramji Dwivedi Ganesh R. Sinha |
| author_sort | Silvia Ullo |
| collection | DOAJ |
| description | This article presents a novel method of landslide detection by exploiting the Mask R-CNN capability of identifying an object layout by using a pixel-based segmentation, along with transfer learning used to train the proposed model. A data set of 160 elements is created containing landslide and nonlandslide images. The proposed method consists of three steps: augmenting training image samples to increase the volume of the training data; fine-tuning with limited image samples; and performance evaluation of the algorithm in terms of precision, recall, and F1 measure, on the considered landslide images, by adopting ResNet-50 and 101 as backbone models. The experimental results are quite encouraging as the proposed method achieves precision equals to 1.00, recall 0.93, and F1 measure 0.97, when ResNet-101 is used as backbone model, and with a low number of landslide photographs used as training samples. The proposed algorithm can be potentially useful for land-use planners and policymakers of hilly areas where intermittent slope deformations necessitate landslide detection as a prerequisite before planning. |
| format | Article |
| id | doaj-art-5b3cd84c63cb41dfa8060821261fcb61 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-5b3cd84c63cb41dfa8060821261fcb612025-08-25T23:00:18ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352021-01-01143799381010.1109/JSTARS.2021.30649819373966A New Mask R-CNN-Based Method for Improved Landslide DetectionSilvia Ullo0https://orcid.org/0000-0001-6294-0581Amrita Mohan1Alessandro Sebastianelli2https://orcid.org/0000-0002-9252-907XShaik Ahamed3Basant Kumar4Ramji Dwivedi5https://orcid.org/0000-0002-9935-1710Ganesh R. Sinha6Department of Engineering, University of Sannio, Benevento, ItalyGIS Cell, Motilal Nehru National Institute of Technology, Prayagraj, IndiaDepartment of Engineering, University of Sannio, Benevento, ItalyMotilal Nehru National Institute of Technology, Prayagraj, IndiaMotilal Nehru National Institute of Technology, Prayagraj, IndiaGIS Cell, Motilal Nehru National Institute of Technology, Prayagraj, IndiaMyanmar Institute of Information Technology, Mandalay, MyanmarThis article presents a novel method of landslide detection by exploiting the Mask R-CNN capability of identifying an object layout by using a pixel-based segmentation, along with transfer learning used to train the proposed model. A data set of 160 elements is created containing landslide and nonlandslide images. The proposed method consists of three steps: augmenting training image samples to increase the volume of the training data; fine-tuning with limited image samples; and performance evaluation of the algorithm in terms of precision, recall, and F1 measure, on the considered landslide images, by adopting ResNet-50 and 101 as backbone models. The experimental results are quite encouraging as the proposed method achieves precision equals to 1.00, recall 0.93, and F1 measure 0.97, when ResNet-101 is used as backbone model, and with a low number of landslide photographs used as training samples. The proposed algorithm can be potentially useful for land-use planners and policymakers of hilly areas where intermittent slope deformations necessitate landslide detection as a prerequisite before planning.https://ieeexplore.ieee.org/document/9373966/Convolutional neural networks (CNNS)global positioning system (GPS)landslide detectionMask R-CNNregion based convolutional neural networks (R-CNN)terrestrial laser scanning (TLS) |
| spellingShingle | Silvia Ullo Amrita Mohan Alessandro Sebastianelli Shaik Ahamed Basant Kumar Ramji Dwivedi Ganesh R. Sinha A New Mask R-CNN-Based Method for Improved Landslide Detection IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural networks (CNNS) global positioning system (GPS) landslide detection Mask R-CNN region based convolutional neural networks (R-CNN) terrestrial laser scanning (TLS) |
| title | A New Mask R-CNN-Based Method for Improved Landslide Detection |
| title_full | A New Mask R-CNN-Based Method for Improved Landslide Detection |
| title_fullStr | A New Mask R-CNN-Based Method for Improved Landslide Detection |
| title_full_unstemmed | A New Mask R-CNN-Based Method for Improved Landslide Detection |
| title_short | A New Mask R-CNN-Based Method for Improved Landslide Detection |
| title_sort | new mask r cnn based method for improved landslide detection |
| topic | Convolutional neural networks (CNNS) global positioning system (GPS) landslide detection Mask R-CNN region based convolutional neural networks (R-CNN) terrestrial laser scanning (TLS) |
| url | https://ieeexplore.ieee.org/document/9373966/ |
| work_keys_str_mv | AT silviaullo anewmaskrcnnbasedmethodforimprovedlandslidedetection AT amritamohan anewmaskrcnnbasedmethodforimprovedlandslidedetection AT alessandrosebastianelli anewmaskrcnnbasedmethodforimprovedlandslidedetection AT shaikahamed anewmaskrcnnbasedmethodforimprovedlandslidedetection AT basantkumar anewmaskrcnnbasedmethodforimprovedlandslidedetection AT ramjidwivedi anewmaskrcnnbasedmethodforimprovedlandslidedetection AT ganeshrsinha anewmaskrcnnbasedmethodforimprovedlandslidedetection AT silviaullo newmaskrcnnbasedmethodforimprovedlandslidedetection AT amritamohan newmaskrcnnbasedmethodforimprovedlandslidedetection AT alessandrosebastianelli newmaskrcnnbasedmethodforimprovedlandslidedetection AT shaikahamed newmaskrcnnbasedmethodforimprovedlandslidedetection AT basantkumar newmaskrcnnbasedmethodforimprovedlandslidedetection AT ramjidwivedi newmaskrcnnbasedmethodforimprovedlandslidedetection AT ganeshrsinha newmaskrcnnbasedmethodforimprovedlandslidedetection |