Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers
Efficient and accurate landslide recognition is crucial for disaster prevention and post-disaster rescue efforts. However, compared to machine learning, deep learning approaches currently face challenges such as long model runtimes and inefficiency. To tackle these challenges, we proposed a novel kn...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-06-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/13/3340 |
| _version_ | 1797590924026445824 |
|---|---|
| author | Renxiang Huang Tao Chen |
| author_facet | Renxiang Huang Tao Chen |
| author_sort | Renxiang Huang |
| collection | DOAJ |
| description | Efficient and accurate landslide recognition is crucial for disaster prevention and post-disaster rescue efforts. However, compared to machine learning, deep learning approaches currently face challenges such as long model runtimes and inefficiency. To tackle these challenges, we proposed a novel knowledge distillation network based on Swin-Transformer (Distilled Swin-Transformer, DST) for landslide recognition. We created a new landslide sample database and combined nine landslide influencing factors (LIFs) with remote sensing images (RSIs) to evaluate the performance of DST. Our approach was tested in Zigui County, Hubei Province, China, and our quantitative evaluation showed that the combined RSIs with LIFs improved the performance of the landslide recognition model. Specifically, our model achieved an Overall Accuracy (OA), Precision, Recall, F1-Score (F1), and Kappa that were 0.8381%, 0.6988%, 0.9334%, 0.8301%, and 0.0125 higher, respectively, than when using only RSIs. Compared with the results of other neural networks, namely ResNet50, Swin-Transformer, and DeiT, our proposed deep learning model achieves the best OA (98.1717%), Precision (98.1672%), Recall (98.1667%), F1 (98.1615%), and Kappa (0.9766). DST has the lowest number of FLOPs, which is crucial for improving computational efficiency, especially in landslide recognition applications after geological disasters. Our model requires only 2.83 GFLOPs, which is the lowest among the four models and is 1.8242 GFLOPs, 1.741 GFLOPs, and 2.0284 GFLOPs less than ResNet, Swin, and DeiT, respectively. The proposed method has good applicability in rapid recognition scenarios after geological disasters. |
| first_indexed | 2024-03-11T01:30:44Z |
| format | Article |
| id | doaj.art-d059e3c22217453abaea89072024338d |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T01:30:44Z |
| publishDate | 2023-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-d059e3c22217453abaea89072024338d2023-11-18T17:24:50ZengMDPI AGRemote Sensing2072-42922023-06-011513334010.3390/rs15133340Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved TransformersRenxiang Huang0Tao Chen1School of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaSchool of Geophysics and Geomatics, China University of Geosciences, Wuhan 430074, ChinaEfficient and accurate landslide recognition is crucial for disaster prevention and post-disaster rescue efforts. However, compared to machine learning, deep learning approaches currently face challenges such as long model runtimes and inefficiency. To tackle these challenges, we proposed a novel knowledge distillation network based on Swin-Transformer (Distilled Swin-Transformer, DST) for landslide recognition. We created a new landslide sample database and combined nine landslide influencing factors (LIFs) with remote sensing images (RSIs) to evaluate the performance of DST. Our approach was tested in Zigui County, Hubei Province, China, and our quantitative evaluation showed that the combined RSIs with LIFs improved the performance of the landslide recognition model. Specifically, our model achieved an Overall Accuracy (OA), Precision, Recall, F1-Score (F1), and Kappa that were 0.8381%, 0.6988%, 0.9334%, 0.8301%, and 0.0125 higher, respectively, than when using only RSIs. Compared with the results of other neural networks, namely ResNet50, Swin-Transformer, and DeiT, our proposed deep learning model achieves the best OA (98.1717%), Precision (98.1672%), Recall (98.1667%), F1 (98.1615%), and Kappa (0.9766). DST has the lowest number of FLOPs, which is crucial for improving computational efficiency, especially in landslide recognition applications after geological disasters. Our model requires only 2.83 GFLOPs, which is the lowest among the four models and is 1.8242 GFLOPs, 1.741 GFLOPs, and 2.0284 GFLOPs less than ResNet, Swin, and DeiT, respectively. The proposed method has good applicability in rapid recognition scenarios after geological disasters.https://www.mdpi.com/2072-4292/15/13/3340landslide recognitiondeep learningknowledge distillationefficiency |
| spellingShingle | Renxiang Huang Tao Chen Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers Remote Sensing landslide recognition deep learning knowledge distillation efficiency |
| title | Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers |
| title_full | Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers |
| title_fullStr | Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers |
| title_full_unstemmed | Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers |
| title_short | Landslide Recognition from Multi-Feature Remote Sensing Data Based on Improved Transformers |
| title_sort | landslide recognition from multi feature remote sensing data based on improved transformers |
| topic | landslide recognition deep learning knowledge distillation efficiency |
| url | https://www.mdpi.com/2072-4292/15/13/3340 |
| work_keys_str_mv | AT renxianghuang landsliderecognitionfrommultifeatureremotesensingdatabasedonimprovedtransformers AT taochen landsliderecognitionfrommultifeatureremotesensingdatabasedonimprovedtransformers |