Insulator Defect Detection Based on ML-YOLOv5 Algorithm
To address the challenges of balancing accuracy and speed, as well as the parameters and FLOPs in current insulator defect detection, we propose an enhanced insulator defect detection algorithm, ML-YOLOv5, based on the YOLOv5 network. The backbone module incorporates depthwise separable convolution,...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/24/1/204 |
| _version_ | 1827384427627413504 |
|---|---|
| author | Tong Wang Yidi Zhai Yuhang Li Weihua Wang Guoyong Ye Shaobo Jin |
| author_facet | Tong Wang Yidi Zhai Yuhang Li Weihua Wang Guoyong Ye Shaobo Jin |
| author_sort | Tong Wang |
| collection | DOAJ |
| description | To address the challenges of balancing accuracy and speed, as well as the parameters and FLOPs in current insulator defect detection, we propose an enhanced insulator defect detection algorithm, ML-YOLOv5, based on the YOLOv5 network. The backbone module incorporates depthwise separable convolution, and the feature fusion C3 module is replaced with the improved C2f_DG module. Furthermore, we enhance the feature pyramid network (MFPN) and employ knowledge distillation using YOLOv5m as the teacher model. Experimental results demonstrate that this approach achieved a 46.9% reduction in parameter count and a 43.0% reduction in FLOPs, while maintaining an FPS of 63.6. It exhibited good accuracy and detection speed on both the CPLID and IDID datasets, making it suitable for real-time inspection of high-altitude insulator defects. |
| first_indexed | 2024-03-08T14:57:33Z |
| format | Article |
| id | doaj.art-9a4fbb92dd904683a203000a9823e38b |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-08T14:57:33Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-9a4fbb92dd904683a203000a9823e38b2024-01-10T15:09:02ZengMDPI AGSensors1424-82202023-12-0124120410.3390/s24010204Insulator Defect Detection Based on ML-YOLOv5 AlgorithmTong Wang0Yidi Zhai1Yuhang Li2Weihua Wang3Guoyong Ye4Shaobo Jin5Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaHenan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaHenan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaChina Special Equipment Inspection and Research Institute, Beijing 100029, ChinaHenan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaHenan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, ChinaTo address the challenges of balancing accuracy and speed, as well as the parameters and FLOPs in current insulator defect detection, we propose an enhanced insulator defect detection algorithm, ML-YOLOv5, based on the YOLOv5 network. The backbone module incorporates depthwise separable convolution, and the feature fusion C3 module is replaced with the improved C2f_DG module. Furthermore, we enhance the feature pyramid network (MFPN) and employ knowledge distillation using YOLOv5m as the teacher model. Experimental results demonstrate that this approach achieved a 46.9% reduction in parameter count and a 43.0% reduction in FLOPs, while maintaining an FPS of 63.6. It exhibited good accuracy and detection speed on both the CPLID and IDID datasets, making it suitable for real-time inspection of high-altitude insulator defects.https://www.mdpi.com/1424-8220/24/1/204convolutional neural networksobject detectionfeature fusionattention mechanisms |
| spellingShingle | Tong Wang Yidi Zhai Yuhang Li Weihua Wang Guoyong Ye Shaobo Jin Insulator Defect Detection Based on ML-YOLOv5 Algorithm Sensors convolutional neural networks object detection feature fusion attention mechanisms |
| title | Insulator Defect Detection Based on ML-YOLOv5 Algorithm |
| title_full | Insulator Defect Detection Based on ML-YOLOv5 Algorithm |
| title_fullStr | Insulator Defect Detection Based on ML-YOLOv5 Algorithm |
| title_full_unstemmed | Insulator Defect Detection Based on ML-YOLOv5 Algorithm |
| title_short | Insulator Defect Detection Based on ML-YOLOv5 Algorithm |
| title_sort | insulator defect detection based on ml yolov5 algorithm |
| topic | convolutional neural networks object detection feature fusion attention mechanisms |
| url | https://www.mdpi.com/1424-8220/24/1/204 |
| work_keys_str_mv | AT tongwang insulatordefectdetectionbasedonmlyolov5algorithm AT yidizhai insulatordefectdetectionbasedonmlyolov5algorithm AT yuhangli insulatordefectdetectionbasedonmlyolov5algorithm AT weihuawang insulatordefectdetectionbasedonmlyolov5algorithm AT guoyongye insulatordefectdetectionbasedonmlyolov5algorithm AT shaobojin insulatordefectdetectionbasedonmlyolov5algorithm |