EcoBin is a full-stack, gamified waste management platform designed to help users intelligently classify their waste and earn leaderboard points while preserving the environment.
Featuring stunning, glassmorphism-based UI components and a fully internal AI microservice, this architecture makes instant sustainability decisions effortless.
- Live AR-Style Camera Scanner: Natively tap into mobile and webcam video feeds to instantly snapshot and classify waste in real-time, right from your web browser.
- Deep Learning AI Microservice: Powered by a custom Python Flask API leveraging TensorFlow/Keras to analyze high-resolution images via OpenCV matrix transformations.
- Robust Gamification Engine: Users accumulate "Eco Points" per successful scan, unlocking competitive streaks and live leaderboards to motivate sustainable lifestyles.
- Secure Fallback Taxonomy: Built-in textual taxonomy algorithms written in Java intelligently fall back on exact text matching if the neural network is completely offline.
- Frontend: React.js, Vite, Tailwind CSS, Framer Motion, HTML5 Canvas, HTML5 MediaDevices APIs
- Backend: Java Spring Boot, Spring Security (JWT Tokens), Hibernate/JPA, RESTful APIs
- Database: PostgreSQL (handling raw Base64 data streaming with boundless TEXT column structures)
- AI Microservice: Python, Flask, OpenCV (BGR/RGB inversion), TensorFlow/Keras (
ecobin_model.keras)
Navigate into the EcoBin_Frontend folder and start the Vite development server:
npm install
npm run devTo host the internal classification engine, initialize the Python server on Port 5000:
cd Model
python app.pyExecute the bundled startup sequence for the Spring Boot REST API:
run_backend.batDeveloped by Ayush as a Final Year Project.