Skip to main content
Average Speed Cameras in Europe: How Section Control Works
Insights

Average Speed Cameras in Europe: How Section Control Works

How average speed cameras work, and where Europe's section controls are: counts per country from OpenStreetMap and our camera index, checked against official lists.

Brent van der Heiden10 min read
#average speed cameras#section control#average speed camera map#how do average speed cameras work#trajectcontrole#radar de tramo#radar tronçon#tutor#openstreetmap#speed cameras

An average speed camera measures how long you took to get from one camera to the next, whatever speed you showed as you passed it. That one design choice changes driver behaviour over kilometres of road instead of a few hundred metres, and it is why section control is now mapped in more than half of the European countries we checked.

It also makes the systems hard to see on a map. A section is two points and the road between them, so a camera database that stores dots undercounts it, and national authorities publish their lists in very different formats, if at all.

We counted average speed sections three ways: in OpenStreetMap, in our own camera index, and in the official lists of six countries that publish one. OpenStreetMap held 1,813 average speed objects worldwide on 29 September 2026, and 1,699 of them (94%) are in Europe. This article explains how the systems work, where they are, and how far each source can be trusted.

How average speed cameras work

A section control system has three parts: a camera at the entry point, a camera at the exit point, and software in between.

  1. Entry. A camera reads the number plate of every vehicle and stores it with a timestamp.
  2. Exit. A second camera, a known distance further on, does the same.
  3. Matching. Software pairs the two reads of each plate, divides the distance by the elapsed time, and keeps only the vehicles whose average speed was above the limit.

Spain's traffic authority describes its radares de tramo in the same terms: artificial vision cameras at both ends, a computer that matches equal plates, checks the passage times and computes the average speed. The DGT has run them since 2010 on motorways and on ordinary roads.

The arithmetic is simple, which is part of the appeal. Take a 5 km section with a 110 km/h limit. The shortest legal time through it is 5 ÷ 110 hours, or 2 minutes 44 seconds. A car that takes 2 minutes 30 seconds averaged 120 km/h, whatever speed it showed at either camera. Braking hard in front of the exit camera changes the result by a few seconds at most.

Authorities subtract a measurement tolerance before issuing a fine. In the Netherlands the public prosecution service states the correction as 3 km/h for limits up to 100 km/h and 3% of the measured speed above that.

Because every vehicle's plate is read, section control is also a form of automatic number plate recognition. The same hardware family is covered in our article on ALPR and Flock cameras in OpenStreetMap. What happens to the reads of drivers who kept to the limit is set by national law.

Where the sections are, according to OpenStreetMap

OpenStreetMap maps an average speed section as a relation of type=enforcement with enforcement=average_speed. The relation holds the camera points and the road between them, so one relation normally equals one section in one direction. Taginfo's chronology shows how young this mapping is: 30 objects at the end of 2019, 826 at the end of 2023, 1,411 at the end of 2025 and 1,813 on 28 September 2026.

Counted per country with Geofabrik's taginfo on 28 September 2026, next to the average speed points in our camera index:

Average speed objects mapped in OpenStreetMap, by country
Spain295Poland235Belgium197Lithuania194Czechia182United Kingdom166Netherlands80Italy66Norway65France47Latvia35Belarus32Portugal26Turkey20Germany16
Objects tagged enforcement=average_speed in each Geofabrik country extract, data until 28 September 2026. Mostly relations, one per section and direction. Data © OpenStreetMap contributors, ODbL.
CountryOpenStreetMap objectsCamera index points (2024-05-21)
Spain295121
Poland23586
Belgium197498
Lithuania194124
Czechia182487
United Kingdom166787
Netherlands8055
Italy66152
Norway6542
France4770
Latvia3533
Portugal2619
Germany161
Austria1020

24 of the 45 European extracts contain at least one average speed object. Sweden, Finland, Greece, Hungary and Romania contain none. The United Kingdom is the odd one out in how it is mapped: 84 of its 166 objects are camera points or ways rather than section relations.

The two columns measure different things, so compare them with care:

  • OpenStreetMap counts sections, mostly one relation per direction, drawn by volunteers from signs and official announcements.
  • The camera index counts camera points reported by drivers of the MapMetrics app. A section has at least two, and a multi-lane gantry can produce more.

Where the index is far ahead, as in the United Kingdom (787 points against 166 objects) and Belgium (498 against 197), drivers have reported more camera sites than mappers have drawn sections. Where OpenStreetMap is ahead, as in Spain and Poland, mappers have been thorough. The same pattern appears for fixed speed cameras in our traffic enforcement camera dataset. The whole index is browsable in the camera index and the speed camera map.

Of the 1,554 relations inside a Europe bounding box, 1,242 (80%) carry the enforced maxspeed, and only 201 (13%) have a name. The most common limits are 50 km/h (321 sections), 90 km/h (234) and 80 km/h (188), so section control is as much a tool for towns and main roads as for motorways.

Checked against official lists

Six countries publish something we could compare against. The units differ, so the table states each one.

CountryOfficial figureSource and dateOpenStreetMap
Italy204 Tutor sections, about 1,987 kmPolizia di Stato, 3 August 202666
Spain47 sections on DGT roadsDGT inventory on the National Access Point, published 18 December 2025295
France66 average speed radars (ETVM)Ministry of the Interior list, December 202547
Netherlands31 locations, both directionsOpenbaar Ministerie80
Austria3 sections set by ordinanceTransport ministry10
Germany1 section, B 6 Gleidingen to LaatzenOVG Lüneburg, 13 November 201916

What each comparison shows:

  • Italy is the largest network and the most under-mapped. The Tutor covers 204 sections and about 1,987 km, 178 of them on the Autostrade per l'Italia network. OpenStreetMap has 66. The Polizia di Stato cites a ministerial decree of 13 June 2017 (no. 282) and a decree of 10 June 2026 (prot. no. 125) as the rules for checking the equipment.
  • Spain's gap runs the other way. The DGT's machine-readable inventory lists 47 average speed sections and 690 fixed radar cabins on the roads it manages. Their end points are a median 2.6 km apart in a straight line, and the longest pair is 32 km apart. OpenStreetMap holds 294 relations, 71 of them in Catalonia, which runs its own traffic service and is not in the DGT list. Sections run by cities, and relations drawn per direction, may explain more of the difference. We did not reconcile the lists one by one.
  • France stopped building them. All 66 average speed radars in the ministry's list entered service between 2012 and 2018, most of them in 2013 and 2014. They are 2% of the 3,309 fixed radars on the list.
  • The Netherlands maps each direction. The prosecution service lists 11 motorway and 20 regional road locations, each enforced in both directions. 31 two-way locations make 62 directional sections, against 80 relations in OpenStreetMap, so the open map probably holds some sections that are no longer on the list. We did not check which.
  • Austria's official list is short. The ministry names the Kaisermühlentunnel on the A22 in Vienna, set by an ordinance of 2007 (BGBl. II Nr. 168/2007), and two stretches of the A2. OpenStreetMap maps 10 sections.
  • Germany has one confirmed section. Lower Saxony's police law allows it, and the state's higher administrative court found no serious constitutional objection in November 2019. Only one of the 16 German relations in OpenStreetMap carries that description. The other 15 have no name, and we could not match them to a published section, so treat them as unverified.

The lesson is the same one we found for low emission zones: open map data tracks what is visible and new well, and it keeps what has been switched off. Official lists track legal status, and they rarely say what was removed.

Query the sections yourself

Paste this into overpass-turbo.eu and replace NL with any two-letter country code:

[out:json][timeout:90];
area["ISO3166-1"="NL"][admin_level=2]->.country;
rel["type"="enforcement"]["enforcement"="average_speed"](area.country);
out geom;

The result draws each section with its cameras and the road between them, and exports to GeoJSON. To count instead of draw, replace the last line with out count;. Keep to one country per query, because the public Overpass servers are shared.

To check the arithmetic for a section you drive, measure the distance between the two gantries with the distance calculator and divide by the limit.

Method and limits

  • OpenStreetMap per country: objects tagged enforcement=average_speed in each Geofabrik country extract, from Geofabrik's taginfo, data until 28 September 2026. Extracts include a small border buffer, so a section near a border can count in two countries. Worldwide total and chronology from taginfo.openstreetmap.org.
  • Tags inside Europe: one Overpass query for every average speed relation in the box 34°N to 72°N and 25°W to 45°E, run on 29 September 2026.
  • Camera index: average speed points per country in the MapAtlas Camera Index, snapshot 21 May 2024, aggregate counts only. 3,270 points are assigned to a country.
  • Official lists: fetched on 29 September 2026. The Spanish and French figures are counted from the published files, and the Italian, Dutch, Austrian and German ones are taken from the pages linked in the table.
  • What the numbers do not measure: whether a mapped section is switched on today, how often it enforces, or how many fines it produces. Temporary sections, such as those at roadworks, move and are rarely mapped.

よくある質問

How do average speed cameras work?

Two or more cameras read the number plate of every vehicle at the start and at the end of a stretch of road and record the time of each read. Software matches the plates, divides the known distance by the time taken, and flags vehicles whose average speed was above the limit. Slowing down only near a camera does not help, because the whole section is measured. Spain's traffic authority, the DGT, describes exactly this plate matching for its radares de tramo, which have run since 2010.

Which European countries use average speed cameras?

OpenStreetMap had average speed sections mapped in 24 of the 45 European country extracts we checked on 28 September 2026. Spain had the most with 295 mapped objects, followed by Poland with 235, Belgium with 197, Lithuania with 194, Czechia with 182 and the United Kingdom with 166. Official lists confirm large networks in Italy (204 Tutor sections) and smaller ones in France (66 section radars), the Netherlands (31 locations) and Austria (3 sections). Germany has one, on the B 6 near Hanover.

Is section control legal in Germany?

Yes, in Lower Saxony. The state's police law, section 32(7) of the NPOG, allows it, and the Higher Administrative Court of Lower Saxony ruled on 13 November 2019 (case 12 LC 79/19) that it had no serious constitutional concerns about that provision. The system covers about two kilometres of the B 6 between Gleidingen and Laatzen, south of Hanover. It is the only section of its kind in Germany that we could confirm.

How many average speed cameras are in the MapMetrics camera index?

The MapAtlas Camera Index, built from MapMetrics driver reports, holds 3,270 average speed camera points in its snapshot of 21 May 2024. The United Kingdom has the most with 787, followed by Russia with 593, Belgium with 498 and Czechia with 487. A section has at least two camera points, one at each end, so the number of sections is lower.

役に立ちましたか?シェアしてください。

著者について

Brent van der Heiden

著者

Brent van der Heiden

Co-Founder & CEO at MapAtlas

Brent built MapAtlas out of a conviction that developers deserve location APIs with fair pricing and genuine end-user privacy. Before that he co-founded MapMetrics, a community mapping and navigation app. He writes about geospatial infrastructure, map data, AI search visibility, and how location data powers the products people rely on every day.

すべての記事を見る →
ブログに戻る