中文摘要：

针对电网安全监控中诊断电力线故障的问题，由于电力线出现断股或附着异物故障，影响电力的传输。为此提出了一种基于视频图像的电力线故障诊断方法。由于电力线的视频图像采集是在极其复杂的背景环境下，并且会有不同气候条件的干扰，会影响电力线的检测和最终的诊断。为了解决这个难题，采用了多帧采样和投票策略的线段检测法，就是利用视频图像中电力线的连续性和其方向的一致性将电力线从复杂多变的背景中检测出来，以电力线为中轴线通过灰度统计特征找出线路中疑似故障位置，再根据空间关系特征和类哈尔特征诊断是否出现故障。实验结果证明，改进方法在复杂的背景环境下提高了电力线故障诊断的准确率。

英文摘要：

Aiming at the fault diagnosis of power line in power grid security monitoring, this paper proposes a fault diagnosis method of power line based on video images. The video image acquisition of power lines is in extremely complex background environment, power line is continuous, and its direction is consistent. A line segment detection method based on the voting strategy is used to detect power line. Then, the suspected fault location of the line is found out by gray statistical characteristic. Finally, the fault diagnosis is realized by using spatial relations characteristics and haar-like features. The experimental results show that the algorithm can accurately remove the noise of background, and achieve the fault diagnosis of power line accurately.