A Defect Detection Method for Grading Rings of Transmission Lines Based on Improved YOLOv8

• DOI: 10.3390/en17194767
• Publication Date: 2024-09-24T12:54:13Z
• Authors: Siyu Xiang; Linghao Zhang; Yumin Chen; Peike Du; Yao Wang; Yue Xi; Bing Li; Zhenbing Zhao
• Keywords: Grading (engineering)

Detecting defects in aerial images of grading rings collected by drones poses challenges due to the structural similarity between normal and defective samples. The small visual differences make it hard distinguish extract key features. Additionally, critical defect features often become lost during feature fusion. To address these issues, this paper uses YOLOv8 as baseline model proposes an improved YOLOv8-based method for detecting ring transmission lines. Our approach first integrates CloAttention C2f modules into extraction network, enhancing model’s ability capture identify rings. we incorporate CARAFE fusion network replace original upsampling module, effectively reducing loss information process. Experimental results demonstrate that our achieves average detection accuracy 67.6% defects, marking a 6.8% improvement over model. This significantly enhances effectiveness line