syon-b/sudan-livestock-tracker

Real-time cattle detection and movement prediction system for Sudan using YOLOv8, sklearn, and ML models # Sudan Livestock Tracker 🐄 Real-time cattle detection and movement prediction system for Sudan using computer vision and machine learning. ## 🌟 Features - **Cattle Detection**: YOLOv8-based cow detection from satellite/aerial imagery - **Movement Risk Prediction**: Predicts movement risk based on environmental factors (NDVI, water distance) - **Path Prediction**: Forecasts where cattle will move over 6-24 hours - **Interactive Map**: Real-time visualization with Leaflet.js - **REST API**: Complete FastAPI backend with prediction endpoints ## 🚀 Quick Start ```bash # Install dependencies cd backend && pip install -r requirements.txt # Start API server python app.py # Open frontend open cow-tracker.html ``` Server runs at: `localhost` ## 📊 Models 1. **YOLOv8 Detector** - Finds cows in images 2. **Risk Predictor** - Predicts movement risk (sklearn) 3. **Path Predictor** - Forecasts future positions (trained in 2.5s!) ## 🧪 Test ```bash python backend/test_complete_pipeline.py ``` ## 📡 API - `GET /health` - Health check - `POST /api/movement-risk` - Predict risk - `POST /api/predict` - Predict paths - `GET /docs` - Interactive API docs See full documentation for details. --- **Status**: Production Ready ✅