Design Of Monitor System For Longitudinal Tear Of Conveyor Belt Based On Machine Vision

Conveyor belt is a long-term continuous transportation tool underground in coal mines.Due to the fast running speed of the conveyor belt and other factors,the longitudinal tearing failure of the conveyor belt occurs from time to time,if it cannot be detected and processed quickly,it will cause huge economic losses.Aiming at the problem of longitudinal tearing of underground conveyor belts,a monitoring system for longitudinal tearing of conveyor belts based on machine vision was designed.First study the deep learning theory and the currently popular YOLO series target detection algorithm based on deep learning,in order to meet the data requirements of deep learning algorithms,a longitudinal tear image generation method based on deep convolution to generate adversarial networks is proposed.Using this method,sample type data with high detection accuracy can be generated,and the problem of a small number of longitudinal tear images of the conveyor belt and poor diversity can be solved.Then researched the conveyor belt longitudinal tear detection algorithm,and found that the target detection algorithm YOLOv3-Tiny can quickly extract advanced features and perform detection,which can meet the system's real-time requirements but the accuracy rate is not high.After improving the algorithm,experiments show that the improved algorithm can achieve good results on both the synthetic image training detection model and the real image.Finally,the overall design of the conveyor belt longitudinal tear monitoring system is carried out,and the entire system design is completed through the design and implementation of system hardware and system software modules.The system integrates database,artificial intelligence and other technologies to become a complete system in which the back-end server detects the longitudinal tear of the conveyor belt,and visualizes the display on the visual interface.It also provides timely warning,data image storage,speed recording and other functions.Figure 44;Table 7;Reference 57...