Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI
Deep learning techniques have flourished in recent years and have shown great potential in ground-penetrating radar (GPR) data interpretation. However, obtaining sufficient training data is a great challenge. This paper proposes an intelligent recognition method based on slicing-aided hyper inferenc...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-09-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/18/8470 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850259079174291456 |
|---|---|
| author | Ruimin Chen Ligang Cao Congde Lu Lei Liu |
| author_facet | Ruimin Chen Ligang Cao Congde Lu Lei Liu |
| author_sort | Ruimin Chen |
| collection | DOAJ |
| description | Deep learning techniques have flourished in recent years and have shown great potential in ground-penetrating radar (GPR) data interpretation. However, obtaining sufficient training data is a great challenge. This paper proposes an intelligent recognition method based on slicing-aided hyper inference (SAHI) for GPR images. Firstly, for the problem of insufficient samples of GPR images with structural loose distresses, data augmentation is carried out based on deep convolutional generative adversarial networks (DCGAN). Since distress features occupy fewer pixels on the original image, to allow the model to pay greater attention to the distress features, it is necessary to crop the original images centered on the distress labeling boxes first, and then input the cropped images into the model for training. Then, the YOLOv5 model is used for distress detection and the SAHI framework is used in the training and inference stages. The experimental results show that the detection accuracy is improved by 5.3% after adding the DCGAN-generated images, which verifies the effectiveness of the DCGAN-generated images. The detection accuracy is improved by 10.8% after using the SAHI framework in the training and inference stages, which indicates that SAHI is a key part of improving detection performance, as it significantly improves the ability to recognize distress. |
| format | Article |
| id | doaj-art-2e87cd51ffdd4e428febd62fc12210f3 |
| institution | OA Journals |
| issn | 2076-3417 |
| language | English |
| publishDate | 2024-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-2e87cd51ffdd4e428febd62fc12210f32025-08-20T01:55:58ZengMDPI AGApplied Sciences2076-34172024-09-011418847010.3390/app14188470Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHIRuimin Chen0Ligang Cao1Congde Lu2Lei Liu3Key Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaKey Laboratory of Earth Exploration and Information Techniques of Ministry of Education, Chengdu University of Technology, Chengdu 610059, ChinaDeep learning techniques have flourished in recent years and have shown great potential in ground-penetrating radar (GPR) data interpretation. However, obtaining sufficient training data is a great challenge. This paper proposes an intelligent recognition method based on slicing-aided hyper inference (SAHI) for GPR images. Firstly, for the problem of insufficient samples of GPR images with structural loose distresses, data augmentation is carried out based on deep convolutional generative adversarial networks (DCGAN). Since distress features occupy fewer pixels on the original image, to allow the model to pay greater attention to the distress features, it is necessary to crop the original images centered on the distress labeling boxes first, and then input the cropped images into the model for training. Then, the YOLOv5 model is used for distress detection and the SAHI framework is used in the training and inference stages. The experimental results show that the detection accuracy is improved by 5.3% after adding the DCGAN-generated images, which verifies the effectiveness of the DCGAN-generated images. The detection accuracy is improved by 10.8% after using the SAHI framework in the training and inference stages, which indicates that SAHI is a key part of improving detection performance, as it significantly improves the ability to recognize distress.https://www.mdpi.com/2076-3417/14/18/8470ground-penetrating radar (GPR)deep convolutional generative adversarial networks (DCGAN)YOLOv5slicing-aided hyper inference (SAHI)object detectiondeep learning |
| spellingShingle | Ruimin Chen Ligang Cao Congde Lu Lei Liu Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI Applied Sciences ground-penetrating radar (GPR) deep convolutional generative adversarial networks (DCGAN) YOLOv5 slicing-aided hyper inference (SAHI) object detection deep learning |
| title | Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI |
| title_full | Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI |
| title_fullStr | Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI |
| title_full_unstemmed | Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI |
| title_short | Research on Intelligent Recognition Method of Ground Penetrating Radar Images Based on SAHI |
| title_sort | research on intelligent recognition method of ground penetrating radar images based on sahi |
| topic | ground-penetrating radar (GPR) deep convolutional generative adversarial networks (DCGAN) YOLOv5 slicing-aided hyper inference (SAHI) object detection deep learning |
| url | https://www.mdpi.com/2076-3417/14/18/8470 |
| work_keys_str_mv | AT ruiminchen researchonintelligentrecognitionmethodofgroundpenetratingradarimagesbasedonsahi AT ligangcao researchonintelligentrecognitionmethodofgroundpenetratingradarimagesbasedonsahi AT congdelu researchonintelligentrecognitionmethodofgroundpenetratingradarimagesbasedonsahi AT leiliu researchonintelligentrecognitionmethodofgroundpenetratingradarimagesbasedonsahi |