Jumper Behavior Detection System for Threading Construction Based on Improved RCF Algorithm

Abstract

Overhead lines are an important task of power engineering construction, and they often face the problems of platform tilt and power lines passing through pulley grooves. We have proposed a set of intelligent power line monitoring system, which is mainly used for the detection of the trolley threading and the balance state of the board during the high-voltage line. We propose to use an improved RCF (Richer Convolutional Features for Edge Detection) algorithm to identify the wire on the pulley groove and determine whether it has an out-of-bounds fault. Our algorithm is practical for threading engineering. The wire image can be processed only by using a portable computer. The processing time of each frame is less than 200ms, and the recognition accuracy has reached the level of state-of-the-art. Practice has proved that the system has the advantages of high real-time performance, high image definition, long transmission distance, and convenient installation and disassembly.