The five steps, from trigger to alert
| Step | What happens |
|---|---|
| 1. Detection | The camera detects a vehicle entering its field of view — day or night, in most weather. |
| 2. Plate capture | It photographs the license plate (and often the vehicle) for a readable frame. |
| 3. Recognition | Software reads the plate characters and pulls vehicle descriptors like make, model and color. |
| 4. Record | The plate, timestamp, location and descriptors are logged to the system. |
| 5. Match & alert | The plate is checked against hotlists — stolen vehicles, wanted plates, AMBER alerts. Matches alert the agency. |
Even if there's no match, the record is usually still stored — for a period set by local policy and state law. That "no match still logged" detail is at the heart of the controversy.
What Flock cameras record
- License plate image and characters — the core function
- Timestamp and location — where and when the plate was read
- Vehicle descriptors — make, model, color, distinguishing features
- Optional video — some deployments capture clips around plate events; whether that's enabled is the agency's choice
For the record, Flock cameras do not measure speed — see do Flock cameras track speed.
Where the data goes
The records live in Flock's platform (FlockOS), and access is governed by each agency's data-sharing agreements. That's why a plate logged in one city can surface in investigations elsewhere — reporting has documented out-of-state access to Silicon Valley plate data. How long records are kept varies — see data retention.
Technical walkthroughs worth reading
- Yahoo Tech — "How Do Flock Cameras Work?"
- Jalopnik — How Flock cameras work
- CNET — When Flock comes to town
- City of Bartlesville — Flock FAQ
- City of Leander, TX — Flock FAQ