Online Live Working Safety Monitoring and Early Warning Based on Spatial Cross‐Border Prevention

Abstract

Manually monitoring the safety of personnel working on high‐voltage lines is difficult and unreliable. To overcome this problem, this paper proposes an online high‐voltage live safety monitoring and early warning system based on spatial cross‐border prevention. To develop the proposed system, following analysis of high‐voltage typical live working modes and the minimum approach distance, we divide the safe operation space and use a time‐of‐flight (TOF) depth camera to measure the live working distance in real time. This enables us realize safety early warning by detecting the boundary crossing and intrusion of the safe operation space. Subsequently, we analyze the ambient light and noise that affect the imaging quality of the TOF camera, evaluate the confidence of the depth image pixels, and apply a purpose‐designed adaptive filter guided by a confidence map to repair the depth images. The results of field experiments conducted verify that the proposed system accurate and reliable of this system through field experiments. Compared with manual observations, the system can significantly improve the effectiveness of real‐time live working safety monitoring, and can effectively prevent the occurrence of live working personal injury accidents. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.