AI traffic survey and vehicle counting
Survey-grade counts from video: 14 categories, by direction, in 15-minute slots.
- Status
- BuiltCounter working on real Mombasa Road footage; full survey system specified
- Period
- Sep 2025 → present
- Role
- Sole builder
- Brand
- Tuchopini
A YOLOv8 detection and tracking pipeline that classifies vehicles into the formal 14-category survey scheme (small motorcycle through 6-axle truck trailer), splits them by direction, aggregates into 15-minute slots and writes CSV in the exact column structure transport engineers already use. Built after a client request that came within 24 hours of a demo post. The companion survey system spec adds edge deployment, containerised services and a monitoring UI.
What it does
- YOLOv8 with persistent tracking IDs
- 14-category survey scheme, line-crossing trigger
- Northbound/southbound split, 15-minute aggregation
- Annotated video output for verification
- Edge-first: no internet at the survey site
Matching the incumbent output format is the adoption problem. If an engineer has to change how they work to use the system, they will not use it.
Watch
More ai, vision and iot
Prototype AI, vision and IoT
AI traffic management on Google Cloud
Camera feeds to adaptive signal timing, with a human override.
Google Cloud · Vertex AI · Pub/Sub · Dataflow
Prototype ▶ 1 AI, vision and IoT
Smart traffic lights (ESP8266 + Raspberry Pi)
Two coordinated intersections on commodity hardware, and a YOLO-driven signal demo.
ESP8266 · Raspberry Pi · Python · YOLOv8
AI, vision and IoT
Smart parking occupancy
Two Raspberry Pis, two sensors, one live bay count you can trust.
Raspberry Pi 4 · Python · OpenCV · YOLO