Visual Field Collaboration of Inspection Terminal based on 3D Point Cloud Data of Substation

Abstract

The substation intelligent inspection system utilizes inspection terminals including cameras and robots to complete repetitive work such as daily inspections, infrared temperature measurement, meter reading recognition, etc. Previous work mostly focuses on object detection technology to improve precision and recall of visual defect detection for substation equipment. Meanwhile, parameters of inspection terminals including zero position, rotation angle and magnification are obtained manually by on-site configuration staff. Generally, one substation equipment corresponds to a set of parameters of one terminal. Thus, it fails to make full use of the image information of equipment from different terminals and different angle. To solve the above-mentioned problems, this paper proposes a visual field collaboration of inspection terminal method based on 3D point cloud data of substation. It aims to realize the autonomous collaboration of inspection terminals to obtain multi-angle image information of substation equipment. The efficacy of the proposed method is verified through experimental results of substation.