Nutrivisor 🍎🏋️‍♂️

Nutrivisor is an AI-powered nutrition and fitness web application designed to help users build healthier eating habits through intelligent food recommendations, calorie tracking, diabetic-aware meal analysis, and exercise guidance.

The system combines AI food recognition, nutrition analysis, glycemic index monitoring, personalized meal recommendations, and fitness tracking into a single responsive healthcare platform.

---

Features

🍽 AI Food Detection

• Detects food items using a trained deep learning model
• Displays nutrition information instantly
• Live food scanning support

🩺 Diabetes-Aware Recommendation System

• Supports:

  • Type 1 Diabetes
  • Type 2 Diabetes
  • Prediabetes
  • Gestational Diabetes

• Filters unsafe foods based on:

  • Glycemic Index (GI)
  • Glycemic Load (GL)

📊 Nutrition Tracking

• Tracks:

  • Calories
  • Protein
  • Carbohydrates
  • Fat
  • Fiber

• Daily meal logging system

🥗 Smart Diet Recommendation

• Personalized food recommendations
• Allergy-aware filtering
• Calorie-aware food suggestions
• Meal-time based recommendations

💪 Exercise & Fitness Tracking

• Exercise recommendation system
• Burned calorie tracking
• Net calorie calculation
• Exercise detail pages

📈 Analytics Dashboard

• Daily calorie analysis
• Weekly calorie trends
• Monthly calorie reports
• Interactive charts

👤 User Management

• User authentication system
• Profile management
• Fitness goal tracking
• Body type & activity analysis

🛠 Admin Features

• Add/Edit/Delete foods
• User management
• Exercise management
• Feedback management

---

Tech Stack

Technology	Usage
Python	Backend
Flask	Web Framework
SQLAlchemy	ORM
SQLite	Database
TensorFlow/Keras	AI Model
OpenCV	Camera & Image Processing
HTML/CSS/JavaScript	Frontend
Jinja2	Templating

---

Frontend Technologies Used

• HTML5
• CSS3
• JavaScript
• Jinja2 Templating Engine
• Responsive UI Design

Database Models

• User
• Menu
• Nutrition
• Daily2
• MealLog
• Exercise
• Feed
• LogSession

---

Installation

1. Clone Repository

git clone <repository-url>
cd Nutrivisor

---

2. Create Virtual Environment

python -m venv venv

Activate environment:

Windows

venv\Scripts\activate

Linux/Mac

source venv/bin/activate

---

3. Install Dependencies

pip install -r requirements.txt

Additional Requirement for AI Detection

⚠️ The food detection system requires the trained AI model file:

food_detect_model.hdf5

This file is necessary for the live food recognition feature to function correctly.

Place the model file in the project root directory:

Nutrivisor/
│
├── food_detect_model.hdf5
├── app.py
├── calorie_data.csv
└── ...

Without this model file:

Camera detection will not work Food prediction will fail Live scanning features will be disabled

---

Run Application

python app.py

Application runs on:

http://127.0.0.1:5200

---

📂 Project Structure

Nutri_Final/
│
├── .vscode/
│
├── __pycache__/
│
├── instance/
│
├── model.savedmodel/
│
├── static/
│
├── templates/
│
├── calorie_data.csv
├── calorie_data.xlsx
├── data.xlsx
│
├── newmenu1.db
├── pythonsqlite.db
│
├── python3.py
├── requirements.txt
├── saved_model.pb
├── style.css
│
├── .gitignore
└── README.md

---

Future Improvements

• Cloud deployment
• Mobile application
• Real-time nutrition chatbot
• Voice assistant integration
• Advanced AI meal prediction
• Wearable device integration

---

Contributors

• Surya Krishna H

---

License

This project is developed for educational and research purposes.