The Design And Implementation Of Detection System For Rip Of Conveyor Belt On Machine Vision

Belt conveyor is a major device for cargo transportation in enterprises such as mines, ports and so on. Belts are easily abraded or blocked in the process of transporting different goods, which results in belts being damaged even torn. Once the belt is torn and alarm is not given in time, the entire belt even frame structure will be so severely damaged that it will directly lead to huge economic loss and even threaten the safety of workers. In the paper, a detection system for longitudinal rip of belt based on machine vision with the assistant of a red line laser is designed according to the rule and characteristic of rip of conveyor belt. Laser stripe distortion information is used to detect tearing belt.A fast detection algorithm is also proposed based on high accuracy and high speed of working state of belt conveyor. Firstly, the data is processed based on LIM and linear weighted Ostu algorithm to segmented laser stripe. Then centre line of laser stripe is extracted according to centre extraction algorithm based on Hessian matrix and line tracking. At last, the current condition of conveyor belt is obtained by extracting distortion information of centre line.Detection device for rip of conveyor belt is tested on the BD7-2 belt conveyor in NO.7 branch of Qinhuangdao Port CO. LTD. Devices designed according to structure of conveyor belt are installed and electric cables are laid on the scene. Detection algorithm and software interface is designed by LabVIEW language. It communicates with the control system by OPC protocol and can collect data automatically, that is to say that the whole process can be realized without human intervention.64892 images are collected from conveyor belt scene during system trial operation and the recognition accuracy rate can reach 93.61%. Since it was put into use in November 2015, the detection device has accurately detected the current conditions of conveyor belt and had high resistance ability to the Industrial environment and pollution noise. The system is stable and reliable and had achieved the expected goal.