测控技术2025，Vol.44Issue(4)：35-41,7.DOI:10.19708/j.ckjs.2025.04.302
基于改进YOLOv8n的输电线路异物实时检测研究
Research on Real-Time Detection of Foreign Objects on Transmission Lines Based on Improved YOLOv8n
焦双健 1郭章勇1
作者信息
- 1. 中国海洋大学工程学院,山东青岛 266100
- 折叠
摘要
Abstract
A real-time detection algorithm for foreign objects on transmission lines and towers based on im-proved YOLOv8n is proposed to address the problem of environmental pollution,short circuits,power outages,and fires caused by foreign objects attached to transmission lines.YOLOv8n is taken as the baseline algorithm.In order to improve the detection accuracy of small targets and enhance the detection ability of foreign objects in complex backgrounds,the detection head of the original YOLOv8n algorithm is modified to a dynamic detec-tion head DynamicHead,which enhances the ability of model to extract multi-dimensional features and dynami-cally adjust to different inputs.The non-maximum suppression(NMS)algorithm is improved to Soft-NMS,so the generalization ability and overall detection performance of the model is improved.The experimental results show that the improved algorithm detects mean average precision(mAP)at 95.7％,which is 4.4％higher than the original YOLOv8n algorithm.It achieves high detection accuracy while ensuring real-time detection speed,and has high effectiveness and practicality.关键词
深度学习/YOLOv8n/输电线路异物/动态检测头Key words
deep learning/YOLOv8n/foreign objects on transmission lines/DynamicHead分类
计算机与自动化引用本文复制引用
焦双健,郭章勇..基于改进YOLOv8n的输电线路异物实时检测研究[J].测控技术,2025,44(4):35-41,7.
