Automated deformation detection and interpretation using InSAR data and a multi-task ViT model
Many geological hazards are associated with ground deformations. Prompt and accurate detection and interpretation of ground deformation is therefore vital to geohazard mitigation. Multitemporal Interferometric Synthetic Aperture Radar (MT-InSAR) is an effective geodetic technique for monitoring grou...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-04-01
|
| Series: | International Journal of Applied Earth Observations and Geoinformation |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1569843224001122 |
| _version_ | 1827294102627024896 |
|---|---|
| author | Mahmoud Abdallah Samaa Younis Songbo Wu Xiaoli Ding |
| author_facet | Mahmoud Abdallah Samaa Younis Songbo Wu Xiaoli Ding |
| author_sort | Mahmoud Abdallah |
| collection | DOAJ |
| description | Many geological hazards are associated with ground deformations. Prompt and accurate detection and interpretation of ground deformation is therefore vital to geohazard mitigation. Multitemporal Interferometric Synthetic Aperture Radar (MT-InSAR) is an effective geodetic technique for monitoring ground deformation. However, accurate computation and interpretation of deformation using InSAR are often hindered by various errors and a lack of expert knowledge. We present a new advanced deep learning model based on a multi-task vision transformer (MT-ViT) to automatically detect, locate, and interpret deformation using single interferograms. To address the issue of limited training data in InSAR applications, the proposed model utilizes pre-trained weights from optical images and transfers them to a simulated InSAR dataset. Then real interferograms are used to fine-tune the weights in the network. An overall loss function is designed, which considers the classification and localization losses in the model. The effectiveness of the proposed model is demonstrated using both simulated and real InSAR datasets that contain either coseismic or volcanic deformation. The experimental results from the model are also compared with the state-of-the-art convolutional neural network (CNN) based techniques. The results show significant improvement in both the accuracy of the results and the computational efficiency over the CNN-based approaches. The MT-ViT model achieved 99.4 % classification accuracy, 54.1 % mean intersection over union (IOU), and 0.9 km localization accuracy. A comprehensive evaluation of the hyperparameters in training the MT-ViT model was carried out, which will inform future research in this direction. The research results highlight the promising capabilities of MT-ViT in near real-time deformation monitoring and automated deformation interpretation. |
| first_indexed | 2024-04-24T13:51:17Z |
| format | Article |
| id | doaj.art-3917f1e6c7474aca862e255b1cf7369f |
| institution | Directory Open Access Journal |
| issn | 1569-8432 |
| language | English |
| last_indexed | 2024-04-24T13:51:17Z |
| publishDate | 2024-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | International Journal of Applied Earth Observations and Geoinformation |
| spelling | doaj.art-3917f1e6c7474aca862e255b1cf7369f2024-04-04T05:03:46ZengElsevierInternational Journal of Applied Earth Observations and Geoinformation1569-84322024-04-01128103758Automated deformation detection and interpretation using InSAR data and a multi-task ViT modelMahmoud Abdallah0Samaa Younis1Songbo Wu2Xiaoli Ding3Department of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong Kong, China; Public Works Department, Mansoura University, Mansoura, EgyptPublic Works Department, Mansoura University, Mansoura, EgyptDepartment of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong Kong, China; Research Institution for Land and Space, The Hong Kong Polytechnic University, Hong Kong, ChinaDepartment of Land Surveying and Geo-Informatics, The Hong Kong Polytechnic University, Hong Kong, China; Research Institution for Land and Space, The Hong Kong Polytechnic University, Hong Kong, China; Corresponding author.Many geological hazards are associated with ground deformations. Prompt and accurate detection and interpretation of ground deformation is therefore vital to geohazard mitigation. Multitemporal Interferometric Synthetic Aperture Radar (MT-InSAR) is an effective geodetic technique for monitoring ground deformation. However, accurate computation and interpretation of deformation using InSAR are often hindered by various errors and a lack of expert knowledge. We present a new advanced deep learning model based on a multi-task vision transformer (MT-ViT) to automatically detect, locate, and interpret deformation using single interferograms. To address the issue of limited training data in InSAR applications, the proposed model utilizes pre-trained weights from optical images and transfers them to a simulated InSAR dataset. Then real interferograms are used to fine-tune the weights in the network. An overall loss function is designed, which considers the classification and localization losses in the model. The effectiveness of the proposed model is demonstrated using both simulated and real InSAR datasets that contain either coseismic or volcanic deformation. The experimental results from the model are also compared with the state-of-the-art convolutional neural network (CNN) based techniques. The results show significant improvement in both the accuracy of the results and the computational efficiency over the CNN-based approaches. The MT-ViT model achieved 99.4 % classification accuracy, 54.1 % mean intersection over union (IOU), and 0.9 km localization accuracy. A comprehensive evaluation of the hyperparameters in training the MT-ViT model was carried out, which will inform future research in this direction. The research results highlight the promising capabilities of MT-ViT in near real-time deformation monitoring and automated deformation interpretation.http://www.sciencedirect.com/science/article/pii/S1569843224001122InSARDeformation detection and interpretationMachine learningVision transformer model |
| spellingShingle | Mahmoud Abdallah Samaa Younis Songbo Wu Xiaoli Ding Automated deformation detection and interpretation using InSAR data and a multi-task ViT model International Journal of Applied Earth Observations and Geoinformation InSAR Deformation detection and interpretation Machine learning Vision transformer model |
| title | Automated deformation detection and interpretation using InSAR data and a multi-task ViT model |
| title_full | Automated deformation detection and interpretation using InSAR data and a multi-task ViT model |
| title_fullStr | Automated deformation detection and interpretation using InSAR data and a multi-task ViT model |
| title_full_unstemmed | Automated deformation detection and interpretation using InSAR data and a multi-task ViT model |
| title_short | Automated deformation detection and interpretation using InSAR data and a multi-task ViT model |
| title_sort | automated deformation detection and interpretation using insar data and a multi task vit model |
| topic | InSAR Deformation detection and interpretation Machine learning Vision transformer model |
| url | http://www.sciencedirect.com/science/article/pii/S1569843224001122 |
| work_keys_str_mv | AT mahmoudabdallah automateddeformationdetectionandinterpretationusinginsardataandamultitaskvitmodel AT samaayounis automateddeformationdetectionandinterpretationusinginsardataandamultitaskvitmodel AT songbowu automateddeformationdetectionandinterpretationusinginsardataandamultitaskvitmodel AT xiaoliding automateddeformationdetectionandinterpretationusinginsardataandamultitaskvitmodel |