Skinalyze - Most Popular Vote Track HackKnight Fall '25 Privacy-first skin analysis application powered by AI and Python OpenCV image processing.

┌─────────────────────┐
│  React Frontend     │  Port 5173/5174
│  (Vite + TypeScript)│  Camera capture, UI
└──────────┬──────────┘
           │ HTTP/REST
           ↓
┌─────────────────────┐
│  Node.js Backend    │  Port 8787
│  (Express + TS)     │  API routing, Gemini AI
└──────────┬──────────┘
           │ HTTP/REST
           ↓
┌─────────────────────┐
│  Python Service     │  Port 8788
│  (FastAPI + OpenCV) │  Image processing
└─────────────────────┘

• 🎥 Real-time Camera Feed - Live video capture
• 🔬 AI-Powered Analysis - Gemini Vision API for skin analysis
• 🐍 Python OpenCV Processing - Professional image enhancement
  • CLAHE (Contrast Limited Adaptive Histogram Equalization)
  • Image quality detection (brightness, contrast, sharpness)
  • Gaussian denoising
  • Edge detection for texture analysis
• 💬 Interactive Chat - Ask questions about your analysis
• 🔒 Privacy First - No data storage, text-only results

The Python microservice provides:

• Quality Analysis - Brightness, contrast, sharpness metrics
• Preprocessing - CLAHE enhancement + denoising before AI analysis
• Edge Detection - Canny algorithm
• ROI Extraction - Region of interest cropping
• Contrast Enhancement - Adaptive histogram equalization
• Skin Tone Detection - HSV-based skin detection

For detailed Python service docs, see backend/python_service/README.md

1. Go to https://makersuite.google.com/app/apikey
2. Create API key
3. Copy it

cd backend
copy .env.example .env
notepad .env

Add your key:

GEMINI_API_KEY=your_key_here
PORT=8787

# Backend (Node.js)
cd backend
npm install

# Frontend (React)
cd ../frontend
npm install

# Python OpenCV Service
cd ../backend/python_service
pip install -r requirements.txt

Terminal 1 - Python OpenCV Service:

cd backend/python_service
python app.py

→ Runs on http://localhost:8788

Terminal 2 - Node.js Backend:

cd backend
npm run dev

→ Runs on http://localhost:8787

Terminal 3 - React Frontend:

cd frontend
npm run dev

→ Runs on http://localhost:5173 or 5174

Go to http://localhost:5173 (or 5174)

1. Click Start → camera opens
2. Click Analyze → image is captured
3. Backend: Calls Python service for OpenCV preprocessing
   • Quality check (brightness, contrast, sharpness)
   • CLAHE enhancement + denoising
4. Backend: Sends enhanced image to Gemini AI
5. View results: summary, categories, risk level, next steps
6. Click End → camera stops

• Frontend: Vite + React + TypeScript
• Backend: Express + TypeScript + Gemini Vision API
• Python Service: FastAPI + opencv-python + NumPy

hack/
├── frontend/    # React app
└── backend/     # API server

Privacy First: No data is stored serverside. Everything is processed in the backend.

todo:

1. Chatbot ✅
2. File Upload ✅
3. Landing page ✅
4. OpenCV ✅
5. prompt tune ✅
6. deploy (optional)
7. Probability ✅
8. Hospital ✅