Nutribot assistant by using deep learning models

This project aims to develop a deep learning-based assistant, Nutribot, for food classification and nutritional information prediction. The Nutribot assists users by classifying food items into predefined categories such as "Apple Pie," "Cannoli," "Cheese Plate," "Cheesecake," "Chicken Wings," "Chocolate Cake," "Deviled Eggs," "Donuts," "French Fries," "Frozen Yogurt," "Ice Cream," and "Macarons." To achieve this, deep learning models, including a custom CNN, SqueezeNet, and VGG16, were implemented for feature extraction and classification tasks. The models were trained to predict nutritional information for each food item, such as "452 Calories/100g, Fat 22.85g, Carbs 47g, Protein 5.7g." The solution is integrated into a user-friendly web application developed using Flask, HTML, CSS, and JavaScript, which allows users to upload images of food and receive detailed nutritional breakdowns. This system can aid in health management by providing accurate nutritional information, promoting healthier food choices, and assisting individuals in tracking their dietary intake.

Keywords: Food Classification, Deep Learning, Custom CNN, SqueezeNet, VGG16, Nutritional Information, Flask, Food Recognition, Health Management, Diet Tracking.