InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements
Abstract InSAR monitoring technology is widely used in investigating landslide hazards. Leveraging object detection algorithms to quickly extract landslide information from Wide-Area InSAR measurements is of great significance. Our InSAR-YOLOv8, an algorithm that automatically detects landslides fro...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-024-84626-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841544725448884224 |
|---|---|
| author | Ruopu Ma Haiyang Yu Xuejie Liu Xinru Yuan Tingting Geng Pengao Li |
| author_facet | Ruopu Ma Haiyang Yu Xuejie Liu Xinru Yuan Tingting Geng Pengao Li |
| author_sort | Ruopu Ma |
| collection | DOAJ |
| description | Abstract InSAR monitoring technology is widely used in investigating landslide hazards. Leveraging object detection algorithms to quickly extract landslide information from Wide-Area InSAR measurements is of great significance. Our InSAR-YOLOv8, an algorithm that automatically detects landslides from InSAR measurements, addresses the low accuracy and suboptimal detection performance of existing network models. In this method, we first design and add a detection head specifically targeting small-scale objects. This improvement enhances the model’s ability to extract features across different scales and strengthens its capability to detect landslides of varying sizes. We also replace the original C2f module with the lighter C2f_Faster module to process information more efficiently, making the model lighter and more efficient. Finally, the SIoU loss function replaces the CIoU loss function to improve the bounding box regression and enhance detection accuracy. Our results show that the proposed algorithm achieves a 97.41% mAP50, a 66.47% mAP50:95, and a 92.06% F1 score on the InSAR landslide dataset, while reducing the number of parameters by 25%. Compared with YOLOv8 and other advanced models (YOLOvX, Faster R-CNN, etc.), our model exhibits distinct advantages and possesses a wider range of potential applications in InSAR measurement for landslide detection. |
| format | Article |
| id | doaj-art-2e7585b7f9d3437f87cc8f1a92274ae2 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-2e7585b7f9d3437f87cc8f1a92274ae22025-01-12T12:18:06ZengNature PortfolioScientific Reports2045-23222025-01-0115112210.1038/s41598-024-84626-3InSAR-YOLOv8 for wide-area landslide detection in InSAR measurementsRuopu Ma0Haiyang Yu1Xuejie Liu2Xinru Yuan3Tingting Geng4Pengao Li5School of Surveying and Land Information Engineering, Henan Polytechnic UniversitySchool of Surveying and Land Information Engineering, Henan Polytechnic UniversitySchool of Surveying and Land Information Engineering, Henan Polytechnic UniversitySchool of Surveying and Land Information Engineering, Henan Polytechnic UniversitySchool of Surveying and Land Information Engineering, Henan Polytechnic UniversitySchool of Surveying and Land Information Engineering, Henan Polytechnic UniversityAbstract InSAR monitoring technology is widely used in investigating landslide hazards. Leveraging object detection algorithms to quickly extract landslide information from Wide-Area InSAR measurements is of great significance. Our InSAR-YOLOv8, an algorithm that automatically detects landslides from InSAR measurements, addresses the low accuracy and suboptimal detection performance of existing network models. In this method, we first design and add a detection head specifically targeting small-scale objects. This improvement enhances the model’s ability to extract features across different scales and strengthens its capability to detect landslides of varying sizes. We also replace the original C2f module with the lighter C2f_Faster module to process information more efficiently, making the model lighter and more efficient. Finally, the SIoU loss function replaces the CIoU loss function to improve the bounding box regression and enhance detection accuracy. Our results show that the proposed algorithm achieves a 97.41% mAP50, a 66.47% mAP50:95, and a 92.06% F1 score on the InSAR landslide dataset, while reducing the number of parameters by 25%. Compared with YOLOv8 and other advanced models (YOLOvX, Faster R-CNN, etc.), our model exhibits distinct advantages and possesses a wider range of potential applications in InSAR measurement for landslide detection.https://doi.org/10.1038/s41598-024-84626-3Improved YOLOv8InSAR measurementsAutomatic detectionLandslide |
| spellingShingle | Ruopu Ma Haiyang Yu Xuejie Liu Xinru Yuan Tingting Geng Pengao Li InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements Scientific Reports Improved YOLOv8 InSAR measurements Automatic detection Landslide |
| title | InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements |
| title_full | InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements |
| title_fullStr | InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements |
| title_full_unstemmed | InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements |
| title_short | InSAR-YOLOv8 for wide-area landslide detection in InSAR measurements |
| title_sort | insar yolov8 for wide area landslide detection in insar measurements |
| topic | Improved YOLOv8 InSAR measurements Automatic detection Landslide |
| url | https://doi.org/10.1038/s41598-024-84626-3 |
| work_keys_str_mv | AT ruopuma insaryolov8forwidearealandslidedetectionininsarmeasurements AT haiyangyu insaryolov8forwidearealandslidedetectionininsarmeasurements AT xuejieliu insaryolov8forwidearealandslidedetectionininsarmeasurements AT xinruyuan insaryolov8forwidearealandslidedetectionininsarmeasurements AT tingtinggeng insaryolov8forwidearealandslidedetectionininsarmeasurements AT pengaoli insaryolov8forwidearealandslidedetectionininsarmeasurements |