Visual system design of tower crane based on improved YOLO

Abstract

In view of the defects of the existing tower crane visual system in Cottonwood, this paper proposes a scheme to connect the video camera and PC to local area network through routers, switches, wireless bridges and other devices. The video data real-time transmission is realized by integrating the Web control package of the camera in the upper computer and accessing the monitor screen through IP address. At the same time, in order to solve the problem of narrow scope and big blind area of traditional direct-viewing camera, a new hook detection algorithm based on improved YOLO is proposed and applied to tower crane safety monitoring system, get the ROI of the picture. The P and T parameters of the camera are calculated and the rotation of the spherical camera is controlled to realize intelligent tracking of the hook.