How k1ethoang/surveillance-camera-system works
Project type and purpose
• This “Surveillance-Camera-System” is a full-stack, web-based surveillance solution composed of two Django services (ai-system and traffic-mngt) plus a camera-stream-simulator.
• ai-system provides camera configuration, live streaming and AI-powered object detection (YOLO models).
• traffic-mngt is a companion service for traffic management workflows, consuming camera data or detection events.
• camera-stream-simulator spins up simulated RTSP/HLS streams from video files for end-to-end testing.
File structure
• Top-level directories:
• ai-system: Django project “ai_system” with app “ai_app”
• traffic-mngt: Django project “traffic_mngt” with app “main_app”
• camera-stream-simulator: Docker Compose config and sample videos
• weights: YOLO model files used by ai-system
• Patterns:
• Django MVC (Models in models.py, Views in views.py, Templates in ai-system/templates)
• Management commands in ai_app/management/commands handle background stream tasks
• camera_process module encapsulates registration, streaming and inference logic
• Config & deployment:
• .env.sample in each service
• requirements.txt for Python dependencies
• docker-compose.yml for simulator
README
• The README only supplies the project title.
• File names and directory structure reveal key components and workflows.
Elements
• Blue for Web UI/API
• Green for background/AI services
• Orange for external systems (cameras, DB)
• Rounded rectangles for services
Detailed guidance
• Finally, include the camera-stream-simulator environment (docker-compose) feeding streams to the processing worker.