Automated bridge crack detection method based on lightweight vision models
Abstract Deep learning-based bridge crack detection methods have advantages over traditional methods. We proposed an automated bridge crack detection method using lightweight vision models. First, our study applied the You Only Look Once 4th version (YOLO v4) (Bochkovskiy et al. in Yolov4: Optimal s...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2022-09-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-022-00876-6 |
| _version_ | 1797840703094521856 |
|---|---|
| author | Jian Zhang Songrong Qian Can Tan |
| author_facet | Jian Zhang Songrong Qian Can Tan |
| author_sort | Jian Zhang |
| collection | DOAJ |
| description | Abstract Deep learning-based bridge crack detection methods have advantages over traditional methods. We proposed an automated bridge crack detection method using lightweight vision models. First, our study applied the You Only Look Once 4th version (YOLO v4) (Bochkovskiy et al. in Yolov4: Optimal speed and accuracy of object detection. arXiv:200410934, 2020) to bridge surface crack detection. Then, to achieve model acceleration, some lightweight networks were used to replace the feature extraction network in YOLO v4, which reduced the parameter numbers and the backbone layers. The lightweight design can reduce the computational overhead of the model, making it convenient to deploy on edge platforms with limited computational power. The experimental results showed that the lightweight network-based bridge crack detection model required significantly less storage space at the expense of a slight reduction in precision. Therefore, an improved YOLO v4 crack detection method was proposed to meet real-time running without sacrificing accuracy. The precision, recall, and F1 score of the proposed crack detection method are 93.96%, 90.12%, and 92%, respectively. And the model only required 23.4 MB of storage space, and its frames per second could reach 140.2 frames. Compared with existing bridge crack detection methods, the proposed method showed precision, speed, and model size advantages. |
| first_indexed | 2024-04-09T16:19:06Z |
| format | Article |
| id | doaj.art-3712df803cba4123aa175e3f3f9183ab |
| institution | Directory Open Access Journal |
| issn | 2199-4536 2198-6053 |
| language | English |
| last_indexed | 2024-04-09T16:19:06Z |
| publishDate | 2022-09-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj.art-3712df803cba4123aa175e3f3f9183ab2023-04-23T11:32:30ZengSpringerComplex & Intelligent Systems2199-45362198-60532022-09-01921639165210.1007/s40747-022-00876-6Automated bridge crack detection method based on lightweight vision modelsJian Zhang0Songrong Qian1Can Tan2School of Mechanical Engineering, Guizhou UniversityState Key Laboratory of Public Big Data, Guizhou UniversitySchool of Mechanical Engineering, Guizhou UniversityAbstract Deep learning-based bridge crack detection methods have advantages over traditional methods. We proposed an automated bridge crack detection method using lightweight vision models. First, our study applied the You Only Look Once 4th version (YOLO v4) (Bochkovskiy et al. in Yolov4: Optimal speed and accuracy of object detection. arXiv:200410934, 2020) to bridge surface crack detection. Then, to achieve model acceleration, some lightweight networks were used to replace the feature extraction network in YOLO v4, which reduced the parameter numbers and the backbone layers. The lightweight design can reduce the computational overhead of the model, making it convenient to deploy on edge platforms with limited computational power. The experimental results showed that the lightweight network-based bridge crack detection model required significantly less storage space at the expense of a slight reduction in precision. Therefore, an improved YOLO v4 crack detection method was proposed to meet real-time running without sacrificing accuracy. The precision, recall, and F1 score of the proposed crack detection method are 93.96%, 90.12%, and 92%, respectively. And the model only required 23.4 MB of storage space, and its frames per second could reach 140.2 frames. Compared with existing bridge crack detection methods, the proposed method showed precision, speed, and model size advantages.https://doi.org/10.1007/s40747-022-00876-6Bridge surfaceCrack detectionLightweight convolutional networksYOLO v4 |
| spellingShingle | Jian Zhang Songrong Qian Can Tan Automated bridge crack detection method based on lightweight vision models Complex & Intelligent Systems Bridge surface Crack detection Lightweight convolutional networks YOLO v4 |
| title | Automated bridge crack detection method based on lightweight vision models |
| title_full | Automated bridge crack detection method based on lightweight vision models |
| title_fullStr | Automated bridge crack detection method based on lightweight vision models |
| title_full_unstemmed | Automated bridge crack detection method based on lightweight vision models |
| title_short | Automated bridge crack detection method based on lightweight vision models |
| title_sort | automated bridge crack detection method based on lightweight vision models |
| topic | Bridge surface Crack detection Lightweight convolutional networks YOLO v4 |
| url | https://doi.org/10.1007/s40747-022-00876-6 |
| work_keys_str_mv | AT jianzhang automatedbridgecrackdetectionmethodbasedonlightweightvisionmodels AT songrongqian automatedbridgecrackdetectionmethodbasedonlightweightvisionmodels AT cantan automatedbridgecrackdetectionmethodbasedonlightweightvisionmodels |