Restricted Zone Monitor

What it does

An intrusion-detection system that watches user-defined restricted zones on an RTSP camera and alerts only when a person crosses into them. Zones are drawn as polygons directly on the live video; every alert stores a timestamped image with the bounding box and zone overlay, and a dashboard tracks active vs resolved alerts and statistics. It's CPU-optimised, so it runs without a GPU.

How it works

1. Step 1. You draw the restricted polygon directly on the camera view (minimum three points) and name it.
2. Step 2. A CPU-optimised YOLOv11 model detects people in every frame at a confidence threshold above 50%.
3. Step 3. When a detected person's footprint enters a zone, the system fires — with a 2-second cooldown so one intrusion doesn't spam alerts.
4. Step 4. Each alert saves a snapshot with box and zone overlay, a timestamp, and a status you can resolve from the web dashboard.

Where it's used

• Hazardous machinery areas
• Secure storage
• Perimeter security
• Platform edges

Tech stack

• Python
• Ultralytics YOLOv11
• Flask
• OpenCV
• RTSP

Build something like this

This project is part of our computer vision development and a fixed-price computer vision pilot. Tell us your use case and we will reply within one business day with a first step, a price and a date.