A Vision-Based Broken Strand Detection Method for a Power-Line Maintenance Robot

Abstract

The broken strand of overhead ground wire (OGW), which is mainly caused by lightning strikes or the vibration of OGW, can lead to serious damage to the power grid system. Power-line maintenance work is generally carried out by specialized workers under extra-high voltage live-line conditions which involve great risks and high labor intensity. In this paper, we present a broken strand detection method which can be practically applied by maintenance robots. This method is mainly implemented in three steps. First, we obtain the region of interest (ROI) from the image acquired by the robot. Second, a histogram of an oriented gradients descriptor vector is calculated to obtain the image gradient feature in ROI. In the third step, we apply a multiclassifier which consists of two support vector machines to classify the wires into normal wire, broken strand malfunction, and obstacles on OGW. Experiment results successfully demonstrate the effectiveness of the proposed method.