Research On Longitudinal Tearing Detection Technology Of Heavy Conveyor Belt Based On Machine Vision

Belt conveyors,as one of the most commonly used ore,coal and port cargo transportation tools,have played a huge role in industrial production.The heavy-duty conveyor belt,as the most expensive part of the cost structure of the belt conveyor,is always longitudinally torn due to the long-term wear of the goods,the inertial impact of the goods and the cutting of the tip of the goods.Once this longitudinal tear occurs,if it cannot be detected in time,it will cause great economic losses and casualties to the enterprise.Aiming at the problem that the longitudinal tearing of the conveyor belt surface is difficult to detect,this paper proposes a machine vision-based heavy-duty conveyor belt longitudinal tearing detection technology.This technology uses a CCD camera in conjunction with a red linear laser device to collect information on the surface of the conveyor belt.Due to the addition of red linear laser equipment,the contrast between the target area and the background area has increased.At the same time,the longitudinal tearing source of the directly detected image is converted into the local distortion of the red laser line,which greatly reduces the difficulty of detection and improves the accuracy and speed of detection.Secondly,using the above-mentioned detection system,a large part of the original image captured belongs to the background area,and the target area that is really useful for judging the tearing result is only a small part.In response to this situation,this paper proposes an improved Otsu threshold segmentation algorithm and a gray barycenter secondary extraction algorithm based on directional templates.Thereby quickly and accurately detecting the distortion area of the red laser line,and then determining whether the conveyor belt is longitudinally torn.This research first went to the mine site to conduct field surveys,recorded various possible environmental factors,and then set up experimental devices in the factory.Various hardware devices and circuits including image acquisition,image transmission,and image processing are designed to ensure the stability and efficiency of data transmission.In addition,the system uses MATLAB as a software development platform to analyze and process the collected images.Finally,the light stripe region segmentation technology based on prior knowledge is combined with the detection algorithm designed in this paper to make the performance of the detection system more excellent.Through many experiments to simulate the longitudinal tear of the conveyor belt under various conditions,the experimental results show that the system can detect the source of longitudinal tear more efficiently and accurately,and basically achieves the expected purpose.