The belt conveyor is an important continuous transportation equipment in the coal production process, and its value accounts for about 15% to 50% of the value of the conveyor. The conveyor belt is the second largest rubber industrial product after the tires. According to reports, China has become the world’s largest conveyor belt producer and consumer. In 2008, the output of the conveyor belt has reached 230 million, accounting for approximately 1 percent of the global conveyor belt market. /3 or so. Due to the harsh operating environment, the conveyor belt is prone to tearing accidents such as scratches, tears, and deviations, which will not only cause economic losses, but also seriously threaten the company's safe production.
Since the 1970s, countries all over the world have begun to explain the tear protection technology, such as embedding method, side legislation, ultrasonic scanning, X-ray inspection. However, these methods all have certain limitations, or they have a simple structure and cannot accurately identify faults, and they are only for a single situation and lack comprehensive detection capabilities. This article will try to introduce the machine vision technology into the tape tear detection, using its characteristics of high precision, high speed, non-contact, reproducible dynamic measurement and so on, to study a new tear detection method. Using this technology to detect the surface fault of the conveyor belt can not only exclude the influence of subjective factors, but also can quantitatively describe the performance indicators of the detection, reduce the labor intensity of manual detection, and improve the detection efficiency and accuracy.
Conveyor inspection is mainly based on the characteristics of the conveyor belt. Due to the need for continuous, high-efficiency, and large-angle transportation of the conveyor belt, high performance is required for the conveyor belt in use. The detection system mainly includes three parts: image acquisition system, image processing system, image analysis system and image recognition system.
First, take a clear picture with a high dynamic range industrial camera, and then send the picture to the image processing system (selecting what kind of industrial camera will directly affect the sharpness and contrast of the image acquisition, which will affect the subsequent image analysis and processing. effect). Secondly, image processing software is used in two stages of image preprocessing and image segmentation to provide ideal target images for subsequent processing. Finally, use an image analysis and recognition system to warn if there are scratches, tears, etc. in the picture. This is directly related to the correctness of the replacement tear test.
The use of machine vision technology to monitor the condition of the conveyor belt can detect the potential failure of the conveyor belt as soon as possible, and stop the spread of the fault in a timely manner, which will help improve the automation level of the company's production. Although on-line monitoring of PU conveyor belt based on machine vision research is still in the experimental research stage, some key technologies have yet to be solved. However, machine vision inspection technology is still the current research hotspot. The use of the latest visual technology for the development of the use of technology is the future research direction of conveyor belt visual inspection technology.