A real-time visual inspection system on Raspberry Pi 5 combining YOLOv6n object detection at ~28 FPS on a Hailo-8L NPU with LiDAR-based proximity sensing — distance overlaid on every bounding box in real time.
Developed as part of an MSc course in AIoT Edge Computing at FH Technikum Wien, this system demonstrates how commodity edge hardware — a Raspberry Pi 5 with a Hailo-8L neural processing unit — can run production-grade object detection at real-time speeds while simultaneously fusing LiDAR range data to provide spatial context on every detection.
The system evolved across three milestones: starting with bare NPU detection and LiDAR fusion, adding a proximity-triggered snapshot system with stateful cooldown, and finishing with face recognition, a Flask web dashboard, and Telegram security alerts for unknown visitors.
YOLOv6n at ~28 FPS on Hailo-8L. Distance from LiDAR overlaid on every bounding box. Real-time LiDAR polar map rendered alongside.
LiDAR monitors entry zones and triggers camera snapshots with stateful cooldown to prevent re-triggering events.
Face recognition (dlib) classifies known vs. unknown visitors. Flask dashboard for live monitoring. Telegram alerts on detections.
All detections logged to rotating CSV files with timestamps and confidence scores for post-analysis.
| COMPONENT | SPECIFICATION |
|---|---|
| Compute | Raspberry Pi 5 (4–8 GB RAM) |
| AI Accelerator | Hailo-8L M.2 neural processing unit |
| LiDAR | Hokuyo URG-04LX-UG01 (USB serial) |
| Camera | Raspberry Pi Camera Module v2/v3 (CSI) |
| Model | YOLOv6n — ~28 FPS on Hailo-8L |
| OS | Raspberry Pi OS 64-bit Bookworm |
| Proximity threshold | Configurable — default 1.5 m |