Development of Power Transmission Line Defects Diagnosis System for UAV Inspection based on Binocular Depth Imaging Technology

摘要

China's transmission lines are widely distributed. The power department spends huge manpower and material resources every year on transmission line patrolling. The development of Unmanned Aerial Vehicle (UAV) technology provides a new inspection method for overhead transmission line patrolling. In the present days, most of the inspection images of drones are based on monocular cameras. Due to the limitations of two-dimensional images, this detection method is not sufficient. Considering the issue that defects of transmission lines are not ideal in the complex natural environment background, in this paper a deep imaging UAV inspection transmission line defect and lurking peril diagnosis system is proposed. This paper introduces the related theory of binocular stereo vision, and describes in detail related image processing techniques such as binocular camera calibration, stereo image correction, stereo matching, and image segmentation. By studying the imaging model of the binocular vision system, a set of automatic transmission line detection platform is established which greatly improve the efficiency of the inspection of the UAV transmission line. Using the system, real-time inspection of transmission lines, intelligent diagnosis of defects is performed which greatly improves the diagnosis efficiency and reduces the manually workload.