Research on foreign body detection method of scraper conveyor based on deep learning
Considering that the foreign bodies such as bolts and pallets are often mixed in the operation of the scraper conveyor,the processing efficiency of the traditional image recognition method is low and the real-time detection is not suitable for the detection of foreign bodies in movement,the foreign body detection method of scraper conveyor based on deep learning technology is studied,as well as the detection algorithm.The foreign body intrusion area is delineated and image preprocessing is carried out,and the YOLOv4 model is used to build a foreign body target detection model,and a large number of image samples are collected on site for calibration and model training.The test results show that the foreign body detection method of scraper conveyor based on deep learning can accurately identify the intruding bolt and pallet and issue an alarm in the complex environment of coal mine.By expanding the data set,the foreign body intrusion detection method can help operators find a variety of foreign bodies in the scraper conveyor in time and reduce the risk of damage to scraper conveyor,crusher and belt conveyor during coal transportation.
scraper conveyorforeign body detectiondeep learningmachine vision
NETL
NSTL
万方数据
