SKU recognizer
AI-powered mobile application for iOS platform that helps sales representatives automatically recognize and track products' presence on shelves using computer vision technology, maintaining detailed reports for each sales point.
Challenge and our solution
When a leading tobacco manufacturer faced a problem of inefficient product placement monitoring across retail locations, the company saw an opportunity to transform the process fundamentally. Brights was entrusted with developing a neural network that could track all SKUs and actualize their stock status using the device's camera. To do this, we turned to artificial intelligence.
Using OpenCV template matching and convolutional neural networks, we created a mobile app enhanced with a system that locates all potential product packs on an image based on a predefined pattern. The selected areas are then analyzed by CNN (convolutional neural networks), trained to identify 50+ types of products (in this particular case, cigarette packs). As a result, we presented the client with a precise recognition system ready for any SKU they needed to track.
By leveraging this AI technology, we significantly enhanced the quality of object recognition, allowing FMCG supervisors to confidently ensure that products are correctly arranged on shelves and that no SKU is missing.
Technologies
Related services
PMI Think Wider
Request a quote.
Thanks for scrolling this far. Let's take the next step. Provide us with a brief description of what you are going to build.