The Detection Algorithm of Broken Wires in Power Lines Based on Grabcut Segmentation

Abstract

Unmanned aerial vehicle (uav) has the advantage of rapid and efficient in power line fault detection. However, the background of power line image taken by uav is complex, and it is difficult to extract the foreground power line for fault detection.Aiming at this problem, we propose an algorithm based on Meanshift superpixel segmentation and Grabcut for further segmentation optimization. At the same time, the RANSAC algorithm is used for denoising to extract accurate power lines. Finally, the power lines are constrained based on their slender characteristics. Compare the width change on the same wire to detect the occurrence of strand breakage in the power line. The experimental results show that the algorithm has a good effect on the detection of broken strands in power lines.