MULTI VISION RECIPE RECOMMENDATION SYSTEM

Abstract

This project uses the Convolutional Neural Network techni and content based filtering for recommendation CNN is that it que for image processing system. The main advantage of using automatically detects the important features without any human supervision. Moreover, content based filtering does not need any data about other users, since the recommendations are specific to the particular user. This makes it easier to scale to a large number of users. The goal of this project is to create image recognition algorithms that can user identify ingredients from images and then recommend recipes based on those ingredients. The recipes are also recommended depending on the user’s BMI (Body Mass Index), which results in healthy recipes for that person. This report examines various techniques for identifying ingredients. The pre-processing step, segmentation, and feature extraction are just a few of the various image processing phases that will be examined and discussed. The output of the algorithms will then be written in the Colab notebook for the backend. The currently build system is a mobile application that performs recommendation based on image recognition and BMI of the user. If the user chooses to process images, the system begins by doing a pre-process on the image after which it recommends recipes based on the images it has recognized. If user selects for BMI, then the system first asks the to enter his/her height and weight. The system then continues by calculating BMI, after which recipes are recommended based on BMI to determine which ones will be healthier for the user. The user can also search for recipes by simply writing the names of ingredients