基于深度学习的输电线路关键部件视觉检测方法的研究进展

Research progress of visual detection methods for transmission line key components based on deep learning

由于电网规模增长,直升机、无人机巡线的大量应用,产生的航拍图像数量剧增,使输电线路关键部件视觉检测与运检人员数量配置的矛盾日益突出。虽然深度学习技术可显著提高目标检测的准确率,但航拍巡线图像背景复杂,关键部件之间的相互遮挡,标注数据量较少等特点,限制了航拍输电线路关键部件视觉检测的工程应用。本文分析了深度学习中目标检测模型的现状,总结了基于深度学习的输电线路关键部件视觉检测方法的研究进展,并指出了构建输电线路关键部件图像数据库、建立专业的输电线路关键部件知识图谱以及将知识图谱与深度模型相融合对输电线路关键部件检测的重要性。

The rapid growth of the power grid and large number of applications of helicopters and unmanned aerial patrol lines have resulted in a dramatic increase in the number of aerial images.The contradiction between the image detection of key components of transmission lines and the number of inspection personnel has become increasingly prominent.Although deep learning can significantly improve the object detection accuracy,the features of complex background the aerial patrol line image,mutual occlusion between the key components and small amount of annotation data limit the engineering applicability of transmission line key components detection.This paper analyses the status of object detection models in deep learning,and summarizes the progress of visual detection methods for transmission line key components based on deep learning.In this paper the importance of constructing of the image database of the key components of transmission lines,establishing knowledge graph of the key components of aerial transmission lines,and combining knowledge graph with deep models are pointed out.