Revolutionizing Retail Stores with Computer Vision and Edge AI: A Novel Shelf Management System

Retail stores often exhibit disorganized features, such as suboptimal store layout, inadequate inventory manage-ment, disarrayed items, and irregularly stocked shelves. These issues can lead to a loss of customers, as well as stock-outs or overstocking, resulting in revenue loss and wastage. Stores find it challenging to attract and retain customers, leading to declining revenues and eventually closures. To improve the competitiveness of stores and enhance the shopping experience for customers, it is imperative to adopt technology and use it to mitigate these issues. To this extent, this research study proposes a novel shelf management system that utilizes computer vision and edge AI (Artificial Intelligence) to help retail staff identify and restock items, as well as to detect and remove misplaced shelf items. Here, the state-of-the-art object detection, image classification, and OCR models are implemented to identify shelf objects and voids. Then, the product-barcode matching algorithm is used to identify misplaced items and alert employees if required. The proposed approach also includes incorporating task scheduling to ensure seamless performance, as well as optimizing our models for edge deployment.