RS-YOLO: An Improved Defect Detection Method for Transmission Line Insulators.

Saved in:

Bibliographic Details
Title:	RS-YOLO: An Improved Defect Detection Method for Transmission Line Insulators.
Authors:	Xia, Zhenglong¹ 59546461@qq.com, Ren, Yingying² 2020241922@jsnu.edu.cn, Li, Can¹ 57862787@qq.com, Zhou, Yangbing² 2020241925@jsnu.edu.cn, Li, Weilong² 15715290276@163.com, Yao, Qiang² 15249728072@163.com
Source:	Engineering Letters. Jul2026, Vol. 34 Issue 7, p2917-2928. 12p.
Subjects:	Electric insulators & insulation, Object recognition (Computer vision), Image processing, Drone surveillance, Artificial intelligence, Cost functions
Abstract:	Recent years have witnessed the widespread application of artificial intelligence (AI) technology in the field of defect detection and early warning for power equipment, with insulator defect detection being directly linked to the reliability and safe operation of transmission systems. However, drone-based insulator inspection still faces challenges such as inconspicuous target features, insufficient small target recognition capability, and the difficulty in balancing detection speed and accuracy. To address these issues, this paper proposes RS-YOLO--an improved algorithm for transmission line insulator defect detection based on the YOLOv11n framework. Firstly, to mitigate feature interference caused by complex backgrounds in insulator images, a Receptive Field Attention Convolution module (RFA-C3k2) is introduced, enhancing the algorithm's perception of multi-scale insulator defects. Secondly, the original neck structure is replaced with SlimNeck -- a lightweight architecture composed of GSConv and VoVGSCSP -- to optimize feature fusion efficiency while reducing computational overhead. Finally, the Weighted Intersection over Union (WIoU) loss function is embedded, leveraging its dynamic focusing mechanism to enhance regression stability and noise resistance in insulator detection. Experimental results on a self-built insulator defect dataset demonstrate that RS-YOLO improves mAP@50 by 3.1% while reducing GFLOPs by 32.9%. The findings validate the effectiveness of the proposed structural improvements in advancing insulator defect detection tasks. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Recent years have witnessed the widespread application of artificial intelligence (AI) technology in the field of defect detection and early warning for power equipment, with insulator defect detection being directly linked to the reliability and safe operation of transmission systems. However, drone-based insulator inspection still faces challenges such as inconspicuous target features, insufficient small target recognition capability, and the difficulty in balancing detection speed and accuracy. To address these issues, this paper proposes RS-YOLO--an improved algorithm for transmission line insulator defect detection based on the YOLOv11n framework. Firstly, to mitigate feature interference caused by complex backgrounds in insulator images, a Receptive Field Attention Convolution module (RFA-C3k2) is introduced, enhancing the algorithm's perception of multi-scale insulator defects. Secondly, the original neck structure is replaced with SlimNeck -- a lightweight architecture composed of GSConv and VoVGSCSP -- to optimize feature fusion efficiency while reducing computational overhead. Finally, the Weighted Intersection over Union (WIoU) loss function is embedded, leveraging its dynamic focusing mechanism to enhance regression stability and noise resistance in insulator detection. Experimental results on a self-built insulator defect dataset demonstrate that RS-YOLO improves mAP@50 by 3.1% while reducing GFLOPs by 32.9%. The findings validate the effectiveness of the proposed structural improvements in advancing insulator defect detection tasks. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X