Abstract: In response to the common problems faced by belt conveyors in the field of longitudinal tear detection, such as high missed detection rate, frequent false alarms, and insufficient intelligence level, we develop an intelligent detection system for longitudinal tearing of strip surfaces that integrates machine vision and line laser technology. Firstly, plan the spatial layout of industrial cameras and line laser emitters to ensure comprehensive and accurate capture of subtle changes on the surface of the conveyor belt; secondly, by introducing efficient image RIO extraction, the data burden in the image processing process is effectively reduced. At the same time, the CLAHE algorithm is used to finely enhance the image, significantly improving the image quality; next, the YOLOv5s deep learning model is used to quickly and accurately identify the key features of longitudinal tearing of the conveyor belt from the enhanced images. Finally, a longitudinal tearing judgment mechanism was proposed, which takes into account the running speed of the belt. By comprehensively analyzing the image recognition results and real-time speed data, it can more accurately determine whether the conveyor belt has truly torn, effectively avoiding misjudgments or omissions caused by a single factor. This system improves the intelligence level of conveyor belt tear detection.