What an ALPR actually reads
Automated license-plate readers are often described as cameras, which undersells them. The interesting part is not the photograph — it is the row in the database it becomes.
An automated license-plate reader — ALPR, sometimes ANPR — is a fixed or vehicle-mounted camera paired with software that converts an image of a passing vehicle into text. Calling it a camera is accurate and slightly misleading: a camera produces a picture, and an ALPR produces a record.
The capture
When a vehicle passes, a typical fixed reader records:
- the plate characters, read by optical character recognition
- a timestamp, usually to the second
- the location of the reader, which is fixed and therefore precise
- vehicle attributes such as color, body type, and sometimes distinguishing features like a roof rack or a bumper sticker
The last item is the one people underestimate. Attribute matching means a vehicle can be searched for without a plate at all, which is why "I have nothing to hide, my plate is not on a list" is an incomplete model of how these systems get used.
The record
The photograph is the least durable part. What persists is a row: this plate, at this corner, at this second. One row is unremarkable. A month of rows from a network of readers is a movement history, and movement histories are legible in a way single sightings are not — they show where you sleep, where you worship, who you visit and how often.
Nothing about that requires a suspect, a warrant, or a case number. The record is created for every vehicle that passes, and the question of who may query it later is a policy question, not a technical one.
Retention is the real variable
Two networks with identical hardware can have completely different privacy profiles depending on one number: how long records are kept.
- Short retention — measured in days. A reader running a 30-day window can answer questions about last month and nothing before it. Most vendor defaults sit in this range.
- Long retention — measured in months or years. The same hardware becomes a historical archive, and queries can reach backwards into a period nobody was under suspicion during.
- Onward sharing — the multiplier. A short retention window matters much less if records are replicated to a national pool where somebody else's retention policy applies.
This is why FlockDetour surfaces retention days and external-sharing counts alongside camera locations. The dot tells you a reader is there; those fields tell you what being read actually costs.
What it does not read
An ALPR does not identify a driver. It identifies a vehicle, and the link from vehicle to person runs through a registration database that the reader itself never touches. That gap is small in practice and large in principle, and most published policy debates live inside it.
FlockDetour holds none of this data. The app never sees your plate, and it does not need to: routing around a reader only requires knowing where the reader is.