TOKYO, JAPAN, March 16, 2018
DOCOMO announced today the launch of an artificial intelligence (AI) engine that quickly, simply and accurately analyzes shelf allocation in stores and warehouses using photos taken with smartphones and other common devices.
The image recognition engine employs DOCOMO's AI technology and constitutes part of NTT Group's corevoTMAI technology. Object-detection technology detects individual items in an image with over 98% accuracy and object-recognition technology identifies specific products with over 95% accuracy by matching them with images stored in a database.
Until now, shelf-analysis technology has required products to be placed in the front row and facing forward to ensure high-precision recognition. DOCOMO's new engine, however, can recognize products on shelves without special arrangement, even when they are packed tightly together. Virtually any product can be recognized immediately if it is registered in the database, meaning that the technology can be used for millions of product types and brands.
DOCOMO expects its new AI engine to be used widely in the distribution and retail industries for product-shelf management and sales analysis. In particular, companies will use the engine to check shelf allocation much faster than through manual methods.
CYBERLINKS became the first company to adopt DOCOMO's engine for an application called TANA SCANTM-AI. The application enables devices such as smartphones to be used to quickly analyze shelf allocation and generate the data required to create a planogram, a visual representation of a store's products.
The new engine was created through DOCOMO's Top GunTM program, under which hybrid teams across the company's R&D and corporate divisions are developing practical and innovative solutions to meet specific needs.
Going forward, DOCOMO expects to collaborate with a variety of companies in applying its new AI engine toward improved productivity in business and society. The initiative is one of many ways DOCOMO is working increasingly with outside partners under its medium-term strategy to 2020.
Illustration Photo: Shelved merchandise (credits: Yuichi Shiraishi / Flickr Creative Commons Attribution 2.0 Generic (CC BY 2.0))