Home / Blog / Computer Vision and AI can make a difference – II

Computer Vision and AI can make a difference – II

Computer vision service allows retailers to use smartphones to take photos of products on store shelves and get immediate feed-back. Image credit: Clobotics

The same company I discussed yesterday in relation to the wind turbine tower inspection, Clobotics, is also engaged in developing a system to help retailers get an up to date inventory of their shelves. Clearly, tags and other IoTs could do the job but these technologies are not easy to deploy and maintain (as you change items on the shelves you need to make sure all IoTs become aware fo the change). Clobotics is leveraging their AI and computer vision technologies to get the same result in a much easier way. They plan to use a smartphone to scan the shelves and their software will identify all products (by detecting their SKU number -Stock Keeping Unit- with a 95% accuracy) and connect to Microsoft BI (Business Intelligence) to create insights on shelf-share ratio, out of stock rate and more. To achieve this high accuracy they trained their computer vision software on consumer packaged goods industry.

They are also considering deploying small digital cameras training the software to recognise products by looking at them (without having to use the SKU code). They are targeting the Coca Cola fridges, over a million of them in China, placing the camera on the fridge door.

By analysing the images they can detect when the door is opened and then closed and comparing the images they can pinpoint what product was taken by the customer (or if products have been added by the shop keeper).

This information, along with information on the fridge temperature and the presence of products from other brands (Coca Cola don’t like having their fridges used to cool other brands…) are sent to a centralised location where data will be analysed. Of course notice will be sent to the shop keeper to warn of stocks depleted, and the need to refill the fridge.

Clobotics aims at being a player in the Digital Transformation, what they call Di-Phy (digitalising the Physical world), using AI and computer vision. These technologies may be cheaper to operate and may deliver better information than other forms of IoT, like tags.

Indeed, AI and computer vision can change the way enterprises monitor and manage “things”. They are easy to install and not invasive (in a way, of course a camera can watch, potentially, much more than “things”). An interesting start up, Moovency, is betting on the uptake of AI&computer vision because, as they say:

  • it is a non-invasive technology
  • it is easy to use
  • has very quick response time
  • can deliver high accuracy
  • can be adapted to several industries

Don’t fool yourself, however, it is a complex technology, and requires top skill to set up a system that works, with quite a bit of research that will still be needed in the coming years. Hats off to those offering it.

About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the Industry Advisory Board within the Future Directions Committee and co-chairs the Digital Reality Initiative. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.