How prajwalay26/medivision works
Project type and purpose
• From the README and file tree, MediVision is a full‑stack web application combining a Flask/Python backend, HTML/CSS/JavaScript frontend, AI/ML model inference, and a cloud database (Firebase). Its purpose is to provide medicine recognition (via image uploads), prescription analysis, text‑to‑speech, multilingual chatbot, and medication reminders.
File structure
• Top‑level files:
• app.py / app2.py – Flask entry points (HTTP routes for image upload, chat, TTS, reminders)
• requirements.txt / packages.txt – Python dependencies (Flask, TensorFlow, OpenCV, Firebase SDK, etc.)
• model files (.h5, .keras) and Jupyter notebooks – AI/ML training and inference artifacts
• ocr_result.txt – sample output from OCR pipeline
• Directories:
• static/ – frontend assets (CSS, JS, images)
• templates/ – HTML templates served by Flask (index.html, medicinepage.html, signup.html, etc.)
• uploads/ – temporary storage for user‑uploaded images and prescriptions
• No separate microservices—backend is a monolith with distinct functional modules (image processing, chatbot, reminders).
Key insights from the README
• Tech stack: Frontend (HTML/CSS/JS), Backend (Flask/Python), Database (Firebase), AI/ML (TensorFlow, OpenCV), External APIs (Speech‑to‑Text, web scraping).
• Core features map to backend modules: medicine identification, TTS engine, chatbot, prescription analyzer, reminder scheduler.