ABSTRACT
The presence of the Internet, Machine Learning, as well as the Internet of Things, have greatly improved many aspects of various industries globally. One of the main benefactors of these technological advancements in the industry, is with retail stores. Beginning from informative websites, online transactions, user-friendly retail applications, and recently, intelligent mobile applications; these have paved the way for retail industries, to improve in various aspects such as process improvement, transaction speed, data accuracy, inventory tracking, customer experience, and others. Organization such as Amazon Go makes use of deep learning, computer vision, and various sensors to provide a “Just Walk Out” shopping experience, allowing for the currently most advanced shopping technology. TPiSHOP, FutureProof Grocery, Future Store, and Regi-Robo Robotic Checkout and Bagging System focuses on allowing self-scanning of items via RFID or QR codes, with some of them having additional self-checkout capabilities. Although these technologies are already available in many countries and many retail industries have started to use them, they are typically known to be very costly to implement and use proprietary technologies, while some of them have not been fully tested nor integrated for use in production. The research investigates an approach to improve retail industries by developing a cost-effective and integrated checkout enhancement platform that implements RFID and image recognition using deep learning, a smart shopping basket, and a real-time integrated tool to improve the customer transaction process.
