Research and Application of Inventory System Based on Visual Recognition

This paper designs a visual recognition based inventory system, which involves: determining the types of inventory targets; Install at least visual recognition cameras and 5G customer terminal equipment in the warehouse; Building a counting model for cigarette trays in storage locations based on the PatMax algorithm and PatQuicks algorithm; Train the counting model of cigarette trays in storage locations; Use a visual recognition camera to take photos of cigarette trays, and use a trained cigarette tray counting model to identify the quantity and brand of cigarettes in the storage tray or non whole tray; Upload the recognition results to the server through the 5G customer terminal device; Analyze and compare the recognition results through the server to obtain statistical results. The research content of this paper can achieve intelligent and unmanned inventory management in flat inventory inventory, and improve the accuracy of inventory management.