GiziMeal
Identify ingredients, understand their nutrition, create balanced meals.
AI Engineer and Project Manager
Team
Web Development
4 Months
Disclaimer
GiziMeal was originally developed as a Capstone Project by team CC26-PSU393 with my role as AI Engineer and Project Manager. This showcase version represents an independent redesign and rebuild of the Frontend and Backend to improve and demonstrate my Web Development capabilities, while the Deep Learning model remains the collaborative work of the team.
Overview
GiziMeal is an interactive web application developed as a capstone project by team CC26-PSU393 under the Healthy Lives and Well-Being theme. The project aims to improve public nutrition literacy through Deep Learning-based food ingredient image classification technology. The app recognizes 15 food ingredient categories from uploaded photos, estimates daily caloric needs using the Mifflin–St Jeor formula based on BMR and TDEE values, and provides balanced meal recommendations aligned with official Indonesian Dietary Reference Intake (AKG) standards alongside a fulfillment score.
Problem Background
- Manual Calorie Counting: Logging and weighing daily food portions manually is time-consuming, causing 80% of users to abandon nutrition tracking.
- Limited Meal Variety: Lack of reference guidelines for building daily meal combinations tailored to specific individual physical needs.
- Low Nutritional Literacy: Shortage of educational resources regarding ideal portion sizes and macronutrient distribution for long-term wellness.
Solution Approach
- Automated Food Image Recognition: Leveraging a convolutional neural network (CNN) to detect ingredients from photos and estimate nutrition instantly.
- Real-time AKG Score Evaluation: Providing balanced menu recommendations tailored to official Indonesian Dietary Reference Intake standards.
- Verified Official Nutrition Data: Combining Kaggle datasets with re-verified local nutritional reference values for maximum accuracy.
- Personalized Caloric Targets: Integrating BMR and TDEE calculations powered by the Mifflin–St Jeor formula based on body physical metrics.
System Workflow
- 1
User captures fresh ingredient photos available in the kitchen using a camera or uploads images from gallery.
- 2
Photos are transmitted to GiziMeal backend for inference processing.
- 3
Deep Learning model analyzes visual surface features and identifies food categories automatically.
- 4
GiziMeal processes ingredient data and prepares balanced menu recommendations with estimated nutrition.
- 5
Detection results, calorie breakdowns, and menu recommendations display on screen in real time.
Tech Stack
Key Features
Food Ingredient Detection and Classification
01Identifies up to 15 food ingredient categories from uploaded images using a Deep Learning model and displays estimated calories.
Balanced Meal Recommendations
02Generates meal recommendations based on detected ingredients in accordance with Indonesian Dietary Reference Intake standards.
BMR and TDEE Calculator
03Calculates personalized daily calorie requirements using the Mifflin–St Jeor method.
Nutrition Information Database
04Provides comprehensive and easily accessible nutritional content information for each food ingredient.
Prediction History Log
05Saves detection history for authenticated users and synchronizes records with the database.
Project Team
Project Gallery
Sign Up Page
User registration form to start a balanced nutrition journey.
Login Page
Secure user authentication sign-in interface.
Home Page
GiziMeal main platform landing page displaying feature highlights and nutrition info.
About Us Page
Information about GiziMeal's vision, mission, and development team.
Food Detection Page
Food ingredient image upload feature for automated AI identification.
Detection Results Page
Analysis breakdown of detected food ingredients and estimated nutritional content.
Menu Recommendation Page
Personalized nutritious meal recipe recommendations based on detected ingredients.
Calorie Calculator Page
Personal BMR and TDEE daily calorie requirement calculation tool.
Nutrition Table Page
Structured directory of food ingredient nutritional facts.
Chatbot Assistant Page
Interactive AI assistant for dietary and nutrition consultations.
Nutrition Reference Page
Official Dietary Recommended Intake guidelines and health references.
FAQ Page
Frequently asked questions regarding GiziMeal platform features.
Account Settings Page
User profile management, password updates, and account preferences.
Detection History Page
Historical records of past food ingredient AI detection analyses.
OTHER PROJECTS
Let's Build Something Together
Reach out if you want to chat about product engineering opportunities, full-time positions, or custom architectural consulting.