A bolt defect detection method for transmission lines based on improved YOLOv5
To solve the problem of bolt defects in unmanned aerial vehicle inspection that are difficult to identify quickly and accurately, this paper proposes a defect detection method based on the improved YOLOv5 anchor mechanism. Firstly, the Normalized Wasserstein distance (NWD) evaluation metric and the...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-03-01
|
| Series: | Frontiers in Energy Research |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fenrg.2024.1269528/full |
| _version_ | 1797269026795159552 |
|---|---|
| author | Hongbo Zou Hongbo Zou Jialun Sun Jialun Sun Ziyong Ye Ziyong Ye Jinlong Yang Jinlong Yang Changhua Yang Changhua Yang Fengyang Li Fengyang Li Li Xiong |
| author_facet | Hongbo Zou Hongbo Zou Jialun Sun Jialun Sun Ziyong Ye Ziyong Ye Jinlong Yang Jinlong Yang Changhua Yang Changhua Yang Fengyang Li Fengyang Li Li Xiong |
| author_sort | Hongbo Zou |
| collection | DOAJ |
| description | To solve the problem of bolt defects in unmanned aerial vehicle inspection that are difficult to identify quickly and accurately, this paper proposes a defect detection method based on the improved YOLOv5 anchor mechanism. Firstly, the Normalized Wasserstein distance (NWD) evaluation metric and the Intersection over Union evaluation metric are combined, and the experiment determines the appropriate weight for this combination. This way, the sensitivity of using IoU alone to small objecet detection anchor box threshold changes was reduced. Furthermore, Convolutional Block Attention Module is included into the head network architecture of yolov5 in order to prioritize significant information and suppress irrelevant features. Omni-dimensional Dynamic Convolution (ODConv) is used to replace convolution in MobileNetv2. The combination module is used as the new backbone of the YOLOv5 model. It simultaneously enhances the model’s capability to extract bolt defect object information, minimizes calculation requirements, and achieves lightweight detection across the entire model. Compared with the original algorithm, the model detection Accuracy Precision (AP) is increased by 30.1%, the mean Accuracy Precision is increased by 30.4%. Other evaluation metrics of the model, such as GFlOPs and Parameters, all decreased slightly. The above results show that the improved algorithm proposed in this paper greatly improves the detection accuracy of the model on the premise of ensuring that the model is as small as possible. |
| first_indexed | 2024-04-25T01:41:50Z |
| format | Article |
| id | doaj.art-2806f740af3e4b05bca59238cf326156 |
| institution | Directory Open Access Journal |
| issn | 2296-598X |
| language | English |
| last_indexed | 2024-04-25T01:41:50Z |
| publishDate | 2024-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Energy Research |
| spelling | doaj.art-2806f740af3e4b05bca59238cf3261562024-03-08T04:19:13ZengFrontiers Media S.A.Frontiers in Energy Research2296-598X2024-03-011210.3389/fenrg.2024.12695281269528A bolt defect detection method for transmission lines based on improved YOLOv5Hongbo Zou0Hongbo Zou1Jialun Sun2Jialun Sun3Ziyong Ye4Ziyong Ye5Jinlong Yang6Jinlong Yang7Changhua Yang8Changhua Yang9Fengyang Li10Fengyang Li11Li Xiong12College of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaCollege of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaCollege of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaCollege of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaCollege of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaCollege of Electrical and New Energy Engineering, China Three Gorges University, Yichang, ChinaHubei Provincial Key Laboratory for Operation and Control of Cascaded Hydropower Station, China Three Gorges University, Yichang, ChinaPower Dispatching and Control Center of Guangxi Power Grid Company, Nanning, ChinaTo solve the problem of bolt defects in unmanned aerial vehicle inspection that are difficult to identify quickly and accurately, this paper proposes a defect detection method based on the improved YOLOv5 anchor mechanism. Firstly, the Normalized Wasserstein distance (NWD) evaluation metric and the Intersection over Union evaluation metric are combined, and the experiment determines the appropriate weight for this combination. This way, the sensitivity of using IoU alone to small objecet detection anchor box threshold changes was reduced. Furthermore, Convolutional Block Attention Module is included into the head network architecture of yolov5 in order to prioritize significant information and suppress irrelevant features. Omni-dimensional Dynamic Convolution (ODConv) is used to replace convolution in MobileNetv2. The combination module is used as the new backbone of the YOLOv5 model. It simultaneously enhances the model’s capability to extract bolt defect object information, minimizes calculation requirements, and achieves lightweight detection across the entire model. Compared with the original algorithm, the model detection Accuracy Precision (AP) is increased by 30.1%, the mean Accuracy Precision is increased by 30.4%. Other evaluation metrics of the model, such as GFlOPs and Parameters, all decreased slightly. The above results show that the improved algorithm proposed in this paper greatly improves the detection accuracy of the model on the premise of ensuring that the model is as small as possible.https://www.frontiersin.org/articles/10.3389/fenrg.2024.1269528/fulltiny object detectiontransmission line bolt defectsevaluation metrics fusionomnidimensional dynamic convolutionanchor-based model |
| spellingShingle | Hongbo Zou Hongbo Zou Jialun Sun Jialun Sun Ziyong Ye Ziyong Ye Jinlong Yang Jinlong Yang Changhua Yang Changhua Yang Fengyang Li Fengyang Li Li Xiong A bolt defect detection method for transmission lines based on improved YOLOv5 Frontiers in Energy Research tiny object detection transmission line bolt defects evaluation metrics fusion omnidimensional dynamic convolution anchor-based model |
| title | A bolt defect detection method for transmission lines based on improved YOLOv5 |
| title_full | A bolt defect detection method for transmission lines based on improved YOLOv5 |
| title_fullStr | A bolt defect detection method for transmission lines based on improved YOLOv5 |
| title_full_unstemmed | A bolt defect detection method for transmission lines based on improved YOLOv5 |
| title_short | A bolt defect detection method for transmission lines based on improved YOLOv5 |
| title_sort | bolt defect detection method for transmission lines based on improved yolov5 |
| topic | tiny object detection transmission line bolt defects evaluation metrics fusion omnidimensional dynamic convolution anchor-based model |
| url | https://www.frontiersin.org/articles/10.3389/fenrg.2024.1269528/full |
| work_keys_str_mv | AT hongbozou aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT hongbozou aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jialunsun aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jialunsun aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT ziyongye aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT ziyongye aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jinlongyang aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jinlongyang aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT changhuayang aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT changhuayang aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT fengyangli aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT fengyangli aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT lixiong aboltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT hongbozou boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT hongbozou boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jialunsun boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jialunsun boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT ziyongye boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT ziyongye boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jinlongyang boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT jinlongyang boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT changhuayang boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT changhuayang boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT fengyangli boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT fengyangli boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 AT lixiong boltdefectdetectionmethodfortransmissionlinesbasedonimprovedyolov5 |