A planogram is the graphical representation of the way a given number of products are positioned within the shelves in a store. The creation of a correct planogram is a fundamental tool for a store's performance: it helps to increase sales and achieve maximum customer satisfactionby reducing out-of-stocks. To this end, this work aims to provide an automatic object recognition based system that allows the operator to verify the correctness of a planogram. For image acquisition, either low-cost battery-powered cameras positioned on the opposite side of the shelf or simply a tablet with a dedicated app can be used. These tools are connected to the cloud where the detection and matching phases are performed. The experimental results come from a real environment.
A deep learning approach for product detection in intelligent retail environment / Pazzaglia, G.; Mameli, M.; Frontoni, E.; Zingaretti, P.; Pietrini, R.; Manco, D.; Placidi, V.. - ELETTRONICO. - 7:(2021). (Intervento presentato al convegno 17th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications, MESA 2021, Held as Part of the ASME 2021 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC-CIE 2021 nel 2021) [10.1115/DETC2021-71462].
A deep learning approach for product detection in intelligent retail environment
Pazzaglia G.;Mameli M.;Frontoni E.;Zingaretti P.;Pietrini R.;Manco D.;
2021-01-01
Abstract
A planogram is the graphical representation of the way a given number of products are positioned within the shelves in a store. The creation of a correct planogram is a fundamental tool for a store's performance: it helps to increase sales and achieve maximum customer satisfactionby reducing out-of-stocks. To this end, this work aims to provide an automatic object recognition based system that allows the operator to verify the correctness of a planogram. For image acquisition, either low-cost battery-powered cameras positioned on the opposite side of the shelf or simply a tablet with a dedicated app can be used. These tools are connected to the cloud where the detection and matching phases are performed. The experimental results come from a real environment.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
