Abstract: Automatic detection method of defects of power line based on visible image in the helicopter inspection system is studied.By analyzing the visible images captured by the camera fixed on the helicopter，defects of power line like line break and foreign matters are detected automatically in real time.The characteristics of the defects are studied，and a method of line object detection using intensity and space information in image is proposed，which can help to identify the power line in the image automatically.All the line objects detected by the proposed method are analyzed to obtain the information about the position，the direction and the width.Based on the information，the objects are classified.By considering the feature of the defects and the information of classifying of the line objects together，line break and foreign matters on power line are detected.The experimental result indicates that the method proposed can detect the power line quite precisely and identify the potential defects like line break and foreign matters and is applied to the helicopter inspection system successfully.

Key words: line break of power line, foreign matters on power line, line feature, line detection, detection of defects

摘要： 研究直升机巡检系统中，基于可见光图像的电力线缺陷诊断方法。对直升机机载系统采集到的可见光图像进行实时分析，自动诊断电力线上可能存在的断股、异物附着缺陷。研究了缺陷在可见光图像中的特征，并提出了一种基于亮度和空间信息的线对象检测方法，自动识别电力线。对线对象进行分析，获取其位置、方向、宽度信息，并进行对象分类。结合缺陷特征和对象与分类信息，设计缺陷诊断流程，诊断断股及异物缺陷。实验证明，该方法对电力线检测准确，能够发现电力线存在的可疑异物和断股缺陷，成功应用于直升机巡检系统中。

关键词: 电力线断股, 电力线上异物, 线特征, 线检测, 缺陷诊断