San Antonio, TX · Military City, USA UEI L58JZMKRCLM5  ·  CAGE 203C1  ·  NAICS 541511  ·  SAM.gov Active

Read this first: these are the ones somebody found. No agency that operates a licence plate reader in Bexar County publishes where they are, so there is no total to compare this against and no coverage rate can be calculated. 540 is a floor. The real number is unknown, and larger.

540plate readers mapped by volunteers in Bexar County
500of them are Flock Safety, 93% of the total
86%have no operator recorded, so who runs them is mostly unknown
210sit outside San Antonio city limits, in the surrounding towns

Nobody official will tell you this

The City of San Antonio publishes 166 open datasets and runs a public GIS service with hundreds of layers. Not one of them is a camera. There is no inventory, no count, and no map of these devices from any agency that operates them. What is on this page came from volunteers walking and driving around and recording what they saw, in OpenStreetMap.

Where the mapped readers are

Flock Safety other makes San Antonio council districts

Every mapped reader is drawn, including the 210 outside the shaded city area: Live Oak, Universal City, Converse, Selma and Windcrest run their own. A dot means a volunteer recorded a device there on some date. It is not a claim that it is still there, and it says nothing about what the device does.

Who operates them is mostly not recorded

This is the single most important gap in the data. Of 540 mapped readers, 464 carry no operator at all. Only 14 distinct operators are named across the rest.

Recorded operators of mapped licence plate readers
OperatorReaders
Live Oak20
Universal City12
Converse7
Lowe's7
Academy Sports + Outdoors5
The Home Depot5
Terrell Hills5
Selma4
City of Windcrest TX3
Windcrest3
VILLAGE GREEN HOMEOWNERS ASSOCIATION2
City of Live Oak1
Not recorded464

No figure on this page supports a sentence of the form "the police department operates N cameras". The record does not say. Note also that not every operator here is a police force: retailers and private businesses appear in this list, and a camera in a shop car park is a different thing from one on a public road.

The operator field is free text, so the same body can appear under more than one spelling. They are listed exactly as tagged rather than merged, because deciding which spellings are the same organisation would be our judgement rather than the record's.

What was mapped, and by whom

Manufacturers of mapped licence plate readers
ManufacturerReadersShare of mapped
Flock Safety50092.6%
(not recorded)213.9%
Motorola Solutions122.2%
Rekor40.7%
Axon Enterprise30.6%

Share of what has been mapped, never a share of what exists.

How current this is

Contributor edits run from 2024-11-13 to 2026-08-28, and 461 of the 540 were touched in the last twelve months. A record is a volunteer observation with a date on it, so the map lags both installations and removals.

Look at it closely

The map loads on demand. Until you press the button, no map service is contacted.

Take the data

Counts as CSV, the full cube as JSON, or through the public API. Source: OpenStreetMap contributors, retrieved 2026-08-28, licensed ODbL 1.0. Unlike the government records behind the other tools on this site, this data is not public domain: anything you derive from it carries the same licence and the same attribution requirement. Cite it as: OpenStreetMap contributors, via brandshyp.org.

What this page is, and is not

An inventory, assembled from the only public source that exists.

An automated licence plate reader photographs every passing number plate, converts it to text, and records it with a time and a place. They are mounted on poles at junctions and on approach roads. Flock Safety is the largest vendor in this market and sells to police departments, homeowner associations and private businesses alike.

This page takes no position on whether they should be there. It answers a narrower question: what is on the public record, and what is missing from it. On this subject the answer is unusually stark. Every other tool on this site audits a government dataset. Here there is no government dataset to audit.

What exists instead is OpenStreetMap, where volunteers, many working through the DeFlock project, have recorded devices they observed. That is a genuine public record and it is checkable: every point can be inspected, edited and disputed by anyone. It is also, unavoidably, incomplete.

Read this before you quote a number from it

This is the weakest evidence base of any tool on this site, and that is worth being blunt about.

There is no denominator, and that is different from the usual complaint

Elsewhere on this site the phrase means there is no traffic count to divide by. Here it is worse: nobody publishes how many readers exist, so there is no total to compare the mapped ones against and no coverage rate can be worked out at all. Every count here is a floor.

Who operates them is mostly unrecorded

Most mapped readers carry no operator tag. No number on this page supports a sentence of the form “the police department runs N cameras”, and nothing here should be read that way. Some tagged operators are private businesses rather than any police force.

Presence only, never behaviour

A record means somebody saw a device at a place on a date. It says nothing about what any camera records, how long anything is kept, who it is shared with, or whether it is even switched on. None of that is in this data and none of it is inferred here.

Plate readers only

The filter is devices tagged as automated plate readers. Ordinary surveillance cameras are excluded by it, including those around federal facilities, which are not published on this page.

Seen one that is not on the map?

The count on this page is a floor, and the only people who can raise it are the ones who drive past these things.

Every reader on the map was put there by a volunteer who noticed one and recorded it. No agency publishes a list to check that against, so the map is not a census and cannot become one. What it can be is less incomplete. If you have seen a reader, the form below records it in a way that keeps every refusal this page already makes: it asks for a part of town rather than an address, it does not ask which way the camera faces, and it does not ask you to guess who operates it.

Seen one that is not on the map? Tell us roughly where and what it looked like. Every answer is a fixed choice: there is no field for an address, and the form does not ask which way the camera faces or who runs it.

Most readers in this county carry a maker's name on the housing. "Not sure" is a perfectly good answer and is recorded as such.

Read by us, never published. Please do not include number plates, phone numbers, email addresses or links: the form refuses them.

What happens to it. A sighting is an unverified statement by somebody we cannot identify, so it is counted separately and never joins the mapped inventory above: that count comes from OpenStreetMap contributors and stays as it is. Nothing is published per part of town until 25 sightings exist. The map remains a floor either way, because no agency publishes a count to check it against.

Why an IT contractor built this

Because the interesting problem is not the map, it is the absence. A city can publish 166 datasets and hundreds of GIS layers and still have nothing at all to say about a network of several hundred cameras. Finding that out took one query against the catalogue and one against the GIS services, and the answer was zero both times.

Working out what a record cannot support is the same discipline whether the record is a permit file, a dispatch log or an empty shelf. See our capabilities or get in touch.

Related free tools

San Antonio Crime Map

Forty-two months of SAPD offences by ZIP with trends and per-capita rates. Official data, with the denominator problems stated.

Aggressive Driving Report

Police call data, the City crash record, and what residents report. Three layers, never summed.

San Antonio Building Permits

Every building permit by type, district and month, published with the coverage rate of every field.