REAL-TIME REMOTE MONITORING AND TRACKING TECHNOLOGY FOR THREE-DIMENSIONAL WAREHOUSE BASED ON DEEP LEARNING AND TARGET DETECTION

In order to realize the dynamic management of goods in the warehouse and improve the overall management efficiency and safety performance of the warehouse, a deep learning and target detection technology for remote real -time monitoring and tracking of the stereo warehouse is proposed. Wireless sensor nodes are used to provide all-round coverage of the three-dimensional warehouse to meet the monitoring needs of various parameters such as cargo status and cargo location, and the image data of the three-dimensional warehouse is trained by deep confidence networks to extract high-quality and distinguished features of the threedimensional warehouse. The hybrid Gaussian model is used to accurately locate and identify the target warehouse, determine the location and trajectory of warehouse objects, and finally introduce a priori and measurement information by virtue of Kalman filtering method to realize real -time monitoring and tracking of the stereo warehouse. The results show that the AUC value of the complex monitoring attributes of the proposed method is as high as 0.93 after the optimization process, and the monitoring accuracy is as high as 98.43%, and the monitoring time is short, and the predicted value of the center of mass coordinates of the warehouse is the same as the real value of the actual coordinates, which indicates that the proposed method can ensure the comprehensiveness and real -time monitoring and tracking, and realize the intelligence and efficiency of the warehouse management.