DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function
Abstract The YOLO series of algorithms have made substantial contributions to the detection of insulator defects in power transmission line operations. However, existing target detection algorithms for the small target detection and low‐quality insulator images encounter difficulties in effectively...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-03-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.13009 |
| _version_ | 1797271958138650624 |
|---|---|
| author | Deao Hu Mei Yu Xianyong Wu Jingbo Hu Yuyang Sheng Yanjing Jiang Chongjing Huang Yuelin Zheng |
| author_facet | Deao Hu Mei Yu Xianyong Wu Jingbo Hu Yuyang Sheng Yanjing Jiang Chongjing Huang Yuelin Zheng |
| author_sort | Deao Hu |
| collection | DOAJ |
| description | Abstract The YOLO series of algorithms have made substantial contributions to the detection of insulator defects in power transmission line operations. However, existing target detection algorithms for the small target detection and low‐quality insulator images encounter difficulties in effectively capturing relevant features, resulting in a higher probability of target loss. To identify and classify defects in the operational state of insulators, an improved YOLOv8 target identification algorithm called DGW‐YOLOv8 is proposed in this paper. The deformable attention backbone of the DGW‐YOLOv8 target identification algorithm is designed by adding the deformable ConvNets v2 module and the global attention mechanism. This addition reduces the feature loss caused by the network feature processing, enhances the sensitivity of the algorithm to small‐scale targets, and reduces the impact caused by the different global positions of the targets. Additionally, to address the problem of low quality of captured images, WIoU v3 is used to replace CIoU in the original YOLOv8 target identification algorithm to optimize the loss function, reduce the degrees of freedom, and improve the network robustness. Experimental results demonstrate that the enhanced YOLOv8 algorithm can achieve an improvement of 2.4% and 5.5% in mAP and mAP50‐95, respectively, compared with the original algorithm. |
| first_indexed | 2024-03-07T14:17:41Z |
| format | Article |
| id | doaj.art-9e2d0c4428dd4f20b81809eb96467c6d |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-03-07T14:17:41Z |
| publishDate | 2024-03-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-9e2d0c4428dd4f20b81809eb96467c6d2024-03-06T11:42:58ZengWileyIET Image Processing1751-96591751-96672024-03-011841096110810.1049/ipr2.13009DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss functionDeao Hu0Mei Yu1Xianyong Wu2Jingbo Hu3Yuyang Sheng4Yanjing Jiang5Chongjing Huang6Yuelin Zheng7Hubei Provincial Collaborative Innovation Center for New Energy Microgrid China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaHubei Provincial Collaborative Innovation Center for New Energy Microgrid China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaHubei Key Laboratory of Intelligent Vision Based Monitoring for Hydroelectric Engineering China Three Gorges University Yichang ChinaAbstract The YOLO series of algorithms have made substantial contributions to the detection of insulator defects in power transmission line operations. However, existing target detection algorithms for the small target detection and low‐quality insulator images encounter difficulties in effectively capturing relevant features, resulting in a higher probability of target loss. To identify and classify defects in the operational state of insulators, an improved YOLOv8 target identification algorithm called DGW‐YOLOv8 is proposed in this paper. The deformable attention backbone of the DGW‐YOLOv8 target identification algorithm is designed by adding the deformable ConvNets v2 module and the global attention mechanism. This addition reduces the feature loss caused by the network feature processing, enhances the sensitivity of the algorithm to small‐scale targets, and reduces the impact caused by the different global positions of the targets. Additionally, to address the problem of low quality of captured images, WIoU v3 is used to replace CIoU in the original YOLOv8 target identification algorithm to optimize the loss function, reduce the degrees of freedom, and improve the network robustness. Experimental results demonstrate that the enhanced YOLOv8 algorithm can achieve an improvement of 2.4% and 5.5% in mAP and mAP50‐95, respectively, compared with the original algorithm.https://doi.org/10.1049/ipr2.13009computer visionimage recognitioninsulatorsobject detection |
| spellingShingle | Deao Hu Mei Yu Xianyong Wu Jingbo Hu Yuyang Sheng Yanjing Jiang Chongjing Huang Yuelin Zheng DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function IET Image Processing computer vision image recognition insulators object detection |
| title | DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function |
| title_full | DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function |
| title_fullStr | DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function |
| title_full_unstemmed | DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function |
| title_short | DGW‐YOLOv8: A small insulator target detection algorithm based on deformable attention backbone and WIoU loss function |
| title_sort | dgw yolov8 a small insulator target detection algorithm based on deformable attention backbone and wiou loss function |
| topic | computer vision image recognition insulators object detection |
| url | https://doi.org/10.1049/ipr2.13009 |
| work_keys_str_mv | AT deaohu dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT meiyu dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT xianyongwu dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT jingbohu dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT yuyangsheng dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT yanjingjiang dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT chongjinghuang dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction AT yuelinzheng dgwyolov8asmallinsulatortargetdetectionalgorithmbasedondeformableattentionbackboneandwioulossfunction |