Hardware LabSoftware LabClient project
AegisEdge Computer Vision Security System
Edge AI video analytics appliance and sensor array delivering real-time perimeter intrusion classification, vehicle tracking, and automated security dispatches.
- detection precision
- 98.6%
- incident alert latency
- < 0.8s
The challenge
Traditional CCTV required continuous manual monitoring by guards, resulting in missed after-hours perimeter breaches and dock safety violations.
What we built
Edge compute appliance running optimized YOLO object detection on RTSP camera streams, microwave radar perimeter triggers, and a cloud dispatch console with instant push alerts.
Tools
- YOLOv8
- OpenCV
- Python
- RTSP
- WebSockets
- React
- Docker
The result
Automated night-shift monitoring across 12 camera zones with sub-second alert delivery and 98.6% classification accuracy.