Abstract: In mining production, crack detection of mining belt conveyors is crucial for ensuring production safety. Belt conveyors are prone to tearing and other malfunctions during long-term load operation. In order to accurately identify the tearing situation of conveyor belts, a conveyor belt tearing judgment and warning system based on laser detection technology has been proposed. The system projects a linear laser beam onto the surface of the conveyor belt to obtain laser stripe images, and then uses multiple image preprocessing techniques, combined with an improved fast robust feature algorithm for image feature extraction, to achieve high-precision detection of surface cracks on the conveyor belt. The experimental results show that the image processing time of the system is stable at around 42 milliseconds, with an average accuracy of 97.51 %, a computer resource utilization rate of only 28.91 %, and a probability of conveyor belt failure of only 0.8. The peak signal-to-noise ratio and structural similarity index of the visualized image output by the system are 38 dB and 0.92, respectively. From this, it can be seen that the method proposed by the research institute can be effectively applied to the detection of conveyor belts in mining areas, reducing belt conveyor failures. This system not only provides a new technology for the detection of conveyor belts in mining belt conveyors, reducing downtime and maintenance costs, but also offers a new approach for conveyor belt detection in other industrial fields.