Scribd
Upload
53 views2 pages

Requirements

The document proposes a recipe recommendation system that uses visual object recognition to identify food ingredients and recommend healthy low-calorie recipes. The system aims to help elders and health-conscious users select recipes by identifying ingredients through photos, text, or voice and searching a database to find matching healthy recipes.

Uploaded by

Chaitanya Singla
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
53 views2 pages

Requirements

The document proposes a recipe recommendation system that uses visual object recognition to identify food ingredients and recommend healthy low-calorie recipes. The system aims to help elders and health-conscious users select recipes by identifying ingredients through photos, text, or voice and searching a database to find matching healthy recipes.

Uploaded by

Chaitanya Singla
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

The proposed system uses real-time visual object recognition of food ingredients and recommends

cooking recipes related to recognized food ingredients. Users can instantly learn about the associated
cooking recipes. The goal of the proposed system is to help elders decide on a cooking recipe in grocery
stores or the kitchen. This task develops a system that allows users to choose healthier foods by
recommending healthier recipes and foods that meet their needs and likes and dislikes. This system is
also beneficial for health-conscious people. This article proposes a recipe recommendation system used
for supervised learning to analyze data using the deep learning method of convolutional neural
networks (CNN).

Main objectives

The goal of the application is to provide a platform where elders find low calories meal recipe for their
favorite food and its nutritional value. The recipe contains several ingredients, cooking procedures, and
categories. The development of low calories meal recommender system using AI technology may lead to
the creation of a global network that will be able to both actively support, and monitor the personalized
supply of nutrients.

Users can able to select the food items by text, voice or by capture/upload image of an ingredient
without any restrictions and using the deep learning neural network (DNN) application to suggest low
calories meal plan. A dietary pattern consistent with current guidelines to consume relatively high
amounts of vegetables, whole grains, fish may be associated with superior nutritional status, quality of
life, and survival in older adults.

Specific objectives

• The objective is to help in the innovation of new dishes and to help people allergic to certain
ingredients by recommending alternate ingredients.

• Nutritional values, e.g., amount of fat or protein included in a recipe or expected by a user.

• Preparation time and difficulty of a meal.

• Variety of low calories’ meal plan, in terms of used ingredients and the category of a meal

The system recognizes the ingredients in the pictures taken by the built-in camera and searches the
online recipe database for recipes that require the ingredients identified. The object detection method
uses a linear kernel SVM with a set of features as image features that use SURFs and color histograms
extracted from multiple images rather than individually, and a one-to-rest strategy as a classifier.

1. Point the smartphone camera at food or enter the ingredient name or by voice.

2. Continuously recognize food ingredients in captured images or entered name.

3. Search the meal plan from the recipe database.


4. Display the low calories meal plan.

• Web scraping

Web scraping is used to extract the data from a website that provides low food calories recipe
recommendations personalized to the individual's taste, various recipes, ingredients, and also the flavor
components of all the ingredients. This data is then stored in the CSV file for further operations.

• Alternative ingredient

Alternative Ingredient recommendation is used to suggest an alternative ingredient if a distinct


ingredient is not present or cannot be used in the recipe. Word2vec model is created to perform this
function and prescribe alternative ingredients.

Tool and Technology (Mobile Application)

Language: Python

Framework: Flask

IDE: Android Studio

Wrap only low calories meal plan (recipes) and store the recipes

Use Google Colab to store the images of the ingredient and do the image classification part

Visual recognition by built in camera and suggest low calories meal plan

Voice recognition of the ingredient and suggest low calories meal plan

Upload a image of the ingredient and suggest low calories meal plan

Using Word2vec model suggest alternative ingredient

So basically 3 methods: by voice, upload an image, and capturing the ingredient

The voice recognition, image classification and the analyzing part for the recipes have to be done in
Python language

It’s enough to use 25 vegetable for the application: Onion, Mushroom, Cabbage, Pumpkin, Ladys finger,
Cauliflower, Tomato, Potato, Raddish, Capsicum, Bitter gourd, Brinjal, Cucumber, Carrot, Beans,
Beetroot, Peas, Lemon, Eggplant, Sweet potato, Corn, Lettuce, Spinach, Olive

Use a background image for the application

You might also like