Grocery items detection using Android studio

Computer science and engineering

In grocery stores, product supply management is critical to employee productivity. Finding the right time to update an item in terms of design/replenishment requires real-time data on item availability. As a result, goods are always available on the shelf when customers need them. This study focuses on product display management in grocery stores to determine the specific products and their quantities on the shelves. Using deep learning (DL) to determine and identify each item, the warehouse manager compares each identified item to the previously created configured item plan. We used You Only Look Once version 5 (YOLOV5) for product detection and used both shape and size features and color features for product detection to reduce false product detection. Experimental results were performed using the dataset. Analysis shows that the proposed approach improved accuracy, precision and recall. Product recognition can shorten defect dates by including color features. It is useful to distinguish identical logos with different colors. You can achieve a functional level 75, reputation level 81 accuracy percentage. We've created a new solution that can reduce checkout and billing times by 50%. What if all the products a customer purchased together were scanned using this algorithm in less than a minute? This algorithm has excellent accuracy, precision and recognition and is considered the best algorithm used in item recognition.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.