This disclosure describes techniques to automatically generate store planograms with the use of network connected cameras. Images are periodically obtained by a camera mounted on a shopping cart while a customer travels through a store. Obtained images are transmitted to a central hub where machine learning techniques such as deep neural networks are utilized to detect item stock and to alert employees if restocking is needed. For example, if an image is received from a camera that indicates a shelf with only a few loaves of bread left, a store employee is automatically notified to restock bread. Additionally, when permitted by the customer, location data is obtained and can be utilized to obtain insight to customer behavior. For example, inefficiencies in store layout can be detected when multiple shopping carts show traversal routes that indicate confusion.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Saha, Samrat, "Automated store planogram generation and maintenance", Technical Disclosure Commons, (August 07, 2019)