Automatic Food Billing System of Indian Food using YOLOv8 model

Automatic food analysis has gained significant interest in the realm of dietary choices as individuals become more aware of their nutritional intake. Particularly in the context of self-service restaurants, this technology not only aids in retrieving information from selected food items but also speeds up the process by solving the congestion problem existing at the cashiers during peak demand periods. This research paper introduces an innovative approach to tackle the challenges associated with automatically analyzing food platters in canteens and restaurants. We have developed a specialized automated billing system tailored specifically for Indian cuisine. Our system leverages computer vision techniques to process food item images and extract vital information like food item names and prices by employing a state of the art object detection algorithm known as YOLO (You Only Look Once). Whether it’s a waiter or a customer, a simple photograph of a food order suffices for our system to extract the necessary details and generate an itemized bill. This eliminates the need for manual item entry, reducing errors and saving time for both customers and businesses. Our approach demonstrates a substantial improvement in food detection, surpassing the state-of-the-art performance on the publicly available IndianFood10 dataset by achieving an impressive 91.9% mAP score. This represents a significant technological advancement in the domain of automatic billing within restaurant settings.