Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO
The insulator defect targets in Unmanned Aerial Vehicle (UAV) images are often small and set against complex backgrounds. Consequently, traditional object detection algorithms commonly struggle to identify these minor defects. To enhance precision and recall in detecting insulator defects, a novel m...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10254232/ |
| _version_ | 1797672892506308608 |
|---|---|
| author | Lu Ding Zhi Qiang Rao Biao Ding Shao Jia Li |
| author_facet | Lu Ding Zhi Qiang Rao Biao Ding Shao Jia Li |
| author_sort | Lu Ding |
| collection | DOAJ |
| description | The insulator defect targets in Unmanned Aerial Vehicle (UAV) images are often small and set against complex backgrounds. Consequently, traditional object detection algorithms commonly struggle to identify these minor defects. To enhance precision and recall in detecting insulator defects, a novel model, GC-YOLO (ghost convolution and centralized feature pyramid -You Only Look Once), based on YOLOv5s, has been introduced. GC-YOLO incorporates the Ghost convolution module in the backbone network, reducing feature redundancy and improving the inference speed of the feature extraction network. Moreover, an attention mechanism based on Coordinate Attention (CA) is integrated at the terminal of the backbone network, aimed at emphasizing the extraction of crucial information from target features. The Explicit Visual Center Block (EVCBlock) module from Centralized Feature Pyramid Network (CFPNet) is introduced in the neck layer to effectively fuse multi-scale features and enhance the feature map’s characterization capability. Furthermore, in order to enhance the precision in detecting small-sized defects, a small object detection head is also added to the detection layer based on CFPNet. Experimental results demonstrate that GC-YOLO achieves a recall of 89.7% and mAP@0.5 of 94.2%, surpassing YOLOv5s by 7% and 6.5%, respectively. The proposed algorithm exhibits superior detection precision in complex scenes, providing a theoretical basis for intelligent and mechanized railway monitoring systems. |
| first_indexed | 2024-03-11T21:36:51Z |
| format | Article |
| id | doaj.art-6e70db3a77484ed4ab24eca7788cff68 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-11T21:36:51Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-6e70db3a77484ed4ab24eca7788cff682023-09-26T23:00:20ZengIEEEIEEE Access2169-35362023-01-011110263510264210.1109/ACCESS.2023.331626610254232Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLOLu Ding0https://orcid.org/0009-0004-8262-8930Zhi Qiang Rao1https://orcid.org/0009-0003-3972-0486Biao Ding2Shao Jia Li3https://orcid.org/0009-0005-6796-8694College of Urban Rail Transit and Logistics, Beijing Union University, Beijing, ChinaCollege of Urban Rail Transit and Logistics, Beijing Union University, Beijing, ChinaCollege of Urban Rail Transit and Logistics, Beijing Union University, Beijing, ChinaCollege of Urban Rail Transit and Logistics, Beijing Union University, Beijing, ChinaThe insulator defect targets in Unmanned Aerial Vehicle (UAV) images are often small and set against complex backgrounds. Consequently, traditional object detection algorithms commonly struggle to identify these minor defects. To enhance precision and recall in detecting insulator defects, a novel model, GC-YOLO (ghost convolution and centralized feature pyramid -You Only Look Once), based on YOLOv5s, has been introduced. GC-YOLO incorporates the Ghost convolution module in the backbone network, reducing feature redundancy and improving the inference speed of the feature extraction network. Moreover, an attention mechanism based on Coordinate Attention (CA) is integrated at the terminal of the backbone network, aimed at emphasizing the extraction of crucial information from target features. The Explicit Visual Center Block (EVCBlock) module from Centralized Feature Pyramid Network (CFPNet) is introduced in the neck layer to effectively fuse multi-scale features and enhance the feature map’s characterization capability. Furthermore, in order to enhance the precision in detecting small-sized defects, a small object detection head is also added to the detection layer based on CFPNet. Experimental results demonstrate that GC-YOLO achieves a recall of 89.7% and mAP@0.5 of 94.2%, surpassing YOLOv5s by 7% and 6.5%, respectively. The proposed algorithm exhibits superior detection precision in complex scenes, providing a theoretical basis for intelligent and mechanized railway monitoring systems.https://ieeexplore.ieee.org/document/10254232/GC-YOLOinsulator-defect detectionCFPNetEVCBlockghost convolution |
| spellingShingle | Lu Ding Zhi Qiang Rao Biao Ding Shao Jia Li Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO IEEE Access GC-YOLO insulator-defect detection CFPNet EVCBlock ghost convolution |
| title | Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO |
| title_full | Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO |
| title_fullStr | Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO |
| title_full_unstemmed | Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO |
| title_short | Research on Defect Detection Method of Railway Transmission Line Insulators Based on GC-YOLO |
| title_sort | research on defect detection method of railway transmission line insulators based on gc yolo |
| topic | GC-YOLO insulator-defect detection CFPNet EVCBlock ghost convolution |
| url | https://ieeexplore.ieee.org/document/10254232/ |
| work_keys_str_mv | AT luding researchondefectdetectionmethodofrailwaytransmissionlineinsulatorsbasedongcyolo AT zhiqiangrao researchondefectdetectionmethodofrailwaytransmissionlineinsulatorsbasedongcyolo AT biaoding researchondefectdetectionmethodofrailwaytransmissionlineinsulatorsbasedongcyolo AT shaojiali researchondefectdetectionmethodofrailwaytransmissionlineinsulatorsbasedongcyolo |