Food Image Recognition for Price Calculation using Convolutional Neural Network

Abstract

This project is attempting to solve the issue of unfair and inconsistent food price being charged in economy rice or mixed rice that widely seen in the café of hawker stall in Malaysia. The main cause of the problem is the absence of standardized price list of the food which causes the pricing of the mixed rice remains unknown. Hence, the authors had decided to propose this project by utilizing convolutional neural network (CNN) algorithm and develop a web application to ease the vendor as well as to provide transparency to the buyer on the food price being charged. CNN model is trained to classify the different types of food. The food price will be stored in a database of the web application in order to calculate the food price with the recognized food in the machine learning model. The outcome of this project is a customized web application for Village 3 Café, UTP with a trained CNN classification model at the backend.