Smart Billing Cart Using RFID, YOLO and Deep Learning for Mall Administration
Abstract
Customers visited the mall to buy the products they needed and pay for them. It is desirable to validate how many goods are sold out before making a customer's bill. When one visits stores to buy something, he or she must exert effort to choose an appropriate item. It is stressful to wait in a queue to be billed after that. Therefore, we suggest a smart cart system be created using RFID to keep track of the products that have been bought as well as online billing transactions. Additionally, the system will recommend goods based on user purchase history from a centralized system. Every item in the store will have an RFID identifier in this system, and every cart will have an RFID reader connected to it. The web recommendation and transaction systems will be centralized. Additionally, there will be an RFID reader for anti-theft purposes at the exit entrance. The suggested system consists of a camera that uses deep learning to identify the item and a load cell that weighs the item that is attached to the shopping cart. When the customer scans the item in front of the camera that is fixed to the cart, the system will produce the bill. Object recognition can be implemented using a variety of techniques. Methods like R-CNN use area proposals to create bounding boxes, which are then used to run classifiers throughout. The duplicates are then removed using a post-processing methodology. R-CNN is a sluggish object recognition technique. We utilize the YOLO paradigm because of this.