Average Speed Cameras in New Zealand: How Point-to-Point Enforcement Works

Average speed cameras are a fundamental shift in how speed enforcement works in New Zealand. Rather than catching drivers at a single point, they measure behaviour over kilometres of road. This guide explains the technology in detail, where the cameras are being deployed, and what the evidence says about their effectiveness.

2026 Average Speed Camera Expansion#

NZTA is rolling out 17 new average speed camera locations across New Zealand through 2026, marking a major expansion of point-to-point enforcement. The first site at Matakana Road, Warkworth became operational in December 2025, with additional locations in Auckland, Waikato, Wellington, Hawke's Bay, and Canterbury following through the year.

For the complete timeline and all confirmed locations, see our 2026 Speed Camera Rollout Guide.

What Are Average Speed Cameras?#

Average speed cameras (also called point-to-point cameras) measure your speed over a distance rather than at a single point. Two or more cameras are placed at known positions along a road. Each camera photographs your vehicle and records the exact time you pass. The system then calculates how long it took you to travel the known distance between cameras and derives your average speed.

That's fundamentally different from a spot speed camera, which only knows how fast you were going at the instant you passed it. Average speed cameras know how fast you drove across an entire corridor.

How the Technology Works#

ANPR: The Core Technology#

The foundation of every average speed camera system is Automatic Number Plate Recognition (ANPR). Each camera unit contains a high-resolution ANPR camera paired with an infrared illuminator. The infrared component is what allows the camera to read number plates clearly regardless of ambient lighting, so the system works equally well at noon, at midnight, in rain, and in fog.

When your vehicle passes a camera, the ANPR system captures an image of your number plate and uses optical character recognition software to convert the plate image into text. Modern systems, including those deployed by NZTA, use deep learning algorithms for plate recognition and achieve approximately 99 percent accuracy.

The cameras also classify vehicle type, distinguishing between cars, trucks, motorcycles, and other vehicles. Different vehicle classes may have different speed limits on the same road, so this classification matters.

The Calculation#

The maths behind average speed calculation is straightforward:

Average Speed = Distance / Time

The distance between camera pairs is precisely surveyed and fixed. The system only needs to measure two timestamps to calculate speed.

Worked Example 1: Highway Corridor#

Cameras are 12 km apart on a 100 km/h road.

• At 100 km/h, the journey should take at least 7 minutes and 12 seconds (7.2 minutes)
• If you pass Camera A at 2:00:00 PM and Camera B at 2:06:30 PM, you took 6 minutes 30 seconds (6.5 minutes)
• Your average speed: 12 km / (6.5/60) hours = 110.8 km/h
• That exceeds the limit and would result in an infringement

Worked Example 2: Slower Road#

Cameras are 5 km apart on an 80 km/h road.

• At 80 km/h, the journey should take at least 3 minutes and 45 seconds
• If you pass Camera A at 9:15:00 AM and Camera B at 9:18:20 AM, you took 3 minutes 20 seconds
• Your average speed: 5 km / (3.333/60) hours = 90.0 km/h
• That exceeds the 80 km/h limit

What Happens When One Camera Fails to Read Your Plate#

No ANPR system is perfect, even at 99 percent accuracy. If the entry camera fails to read your plate but the exit camera reads it successfully, no average speed can be calculated because there's no starting timestamp. No infringement is generated for that trip through the corridor.

The same applies in reverse. If the entry camera captures you but the exit camera fails, no average speed can be determined. The system requires a successful plate read at both ends.

But average speed corridors can have more than two cameras. Some systems use intermediate cameras along the route. If any two cameras in the sequence both successfully read your plate, an average speed can be calculated for the segment between them. That redundancy reduces the impact of any single camera failure.

How the System Handles Lane Changes and Overtaking#

A common concern is whether changing lanes defeats the system. It doesn't. ANPR cameras are positioned to capture plates across all lanes of traffic. Whether you're in the left lane at the entry point and the right lane at the exit, the system matches your plate regardless.

Overtaking is also accounted for. Average speed cameras don't measure your peak speed at any point in the corridor. They measure only your average speed over the entire distance. If you briefly accelerate to 120 km/h to safely overtake a slower vehicle but otherwise drive at 95 km/h, your average speed across the corridor may still be within the limit.

And that's actually one of the key advantages over spot speed cameras. A spot camera would catch the momentary 120 km/h during an overtake. An average speed camera evaluates your sustained driving behaviour, which is a more meaningful measure of risk.

What Happens If You Stop for Fuel or a Rest#

If you pull over within an average speed corridor to refuel, take a break, or stop for any reason, the time you spend stationary is included in the total elapsed time. Stopping will lower your calculated average speed, not raise it.

For example, say you drive a 12 km corridor at an average of 110 km/h (which would normally take about 6 minutes 33 seconds) but stop for 5 minutes at a petrol station partway through. Your total elapsed time becomes approximately 11 minutes 33 seconds. Your calculated average speed would then be about 62 km/h, well within any 100 km/h limit.

In practice, average speed cameras will never penalise you for stopping. The system doesn't distinguish between driving time and stationary time.

Data Encryption and Matching#

In most modern average speed camera systems, including those based on the Jenoptik SPECS technology used widely in the UK, the plate data is encrypted immediately after capture at the entry point. When the same vehicle is detected at the exit point, the system matches the encrypted records. If the calculated average speed is within the limit, the data is discarded. Only records where the speed exceeds the enforcement threshold are retained and forwarded for processing.

NZTA has stated that images where no offence is detected are deleted within minutes at the camera site. Images associated with confirmed offences are retained for three years.

Weather and Visibility#

Average speed cameras using infrared ANPR technology are largely unaffected by weather conditions. The infrared illuminator provides consistent lighting for plate reading regardless of ambient conditions. The cameras automatically switch between day mode (colour images) and night mode (infrared black-and-white images) as needed.

Heavy rain, fog, or snow can theoretically reduce recognition accuracy, but modern systems are designed to maintain high performance in adverse conditions. The 99 percent plate recognition rate cited by NZTA accounts for real-world operating conditions.

Where the Cameras Are Being Deployed#

NZTA selected the 17 average speed camera locations based on three criteria: crash history, traffic volumes, and the proportion of drivers exceeding the posted speed limit. Every chosen corridor has a documented pattern of deaths or serious injuries linked to speed.

North Island Locations#

• Matakana Road, Warkworth (Auckland) - operational since December 2025
• Kahikatea Flat Road (Auckland)
• Pine Valley Road (Auckland)
• Whitford Road (Auckland)
• East Coast Road (Auckland)
• Glenbrook Road (Auckland)
• Kaitaia-Awaroa Road (Northland)
• SH2 Matata (Bay of Plenty)
• SH5 Tumunui (Bay of Plenty)
• SH2 Pokeno (Waikato)
• SH2 Te Hauke (Hawke's Bay)
• SH1 Sanson (Manawatu-Whanganui)
• SH2 Upper Hutt (Wellington)

South Island Locations#

• SH8 Lake Tekapo to Twizel (Canterbury)
• SH1 Allanton to Waihola (Otago)
• SH6 Kingston (Southland)

Construction on southern sites began in late 2025, with remaining sites expected to be completed through mid-2026.

Early Results from Matakana Road#

The first operational site at Matakana Road has provided striking early evidence of effectiveness. Before the cameras were activated, approximately 12 percent of drivers on Matakana Road exceeded the speed limit. After activation, that figure dropped to under 1 percent, meaning more than 99 percent of drivers were compliant.

NZTA has also reported a 60 percent reduction in crashes on Matakana Road in the three months following camera activation. Average speeds on the corridor moved closer to the posted limit (vehicles that were previously travelling well under the limit now travel closer to it, while speeders slowed down), creating a more consistent traffic flow.

International Evidence: The UK Experience#

The United Kingdom has been operating average speed cameras since the early 2000s, making it the most experienced country in the world with this technology. The UK primarily uses the Jenoptik SPECS system, which operates on the same ANPR principles as the cameras being deployed in New Zealand.

By 2015, at least 50 stretches of UK road covering approximately 410 km were permanently monitored by average speed cameras. The accumulated evidence is compelling:

• Research commissioned by the RAC Foundation found a 36 percent reduction in fatal and serious collisions after average speed cameras were introduced
• An overall 16 percent reduction in personal injury collisions of all severities
• The A9 installation in Scotland recorded over a 40 percent reduction in fatal casualties and a 31 percent reduction in combined fatal and serious casualties within three years
• Across Nottinghamshire SPECS installations, killed-or-seriously-injured figures fell by an average of 65 percent
• A Department for Transport evaluation found a 70 percent reduction in speeding and a 6 percent fall in average speeds at fixed camera sites

These results line up with what NZTA is now seeing at Matakana Road. The UK experience suggests that the benefits of average speed cameras hold up over time, unlike spot speed cameras where some drivers learn the exact locations and brake only at those points. I've driven through SPECS zones in the UK myself, and the difference in driver behaviour is noticeable. Everyone just settles into the flow.

Privacy Considerations#

Average speed cameras raise real privacy questions because they photograph every vehicle passing through a corridor, not just those exceeding the speed limit. The ANPR system captures and reads the plate of every vehicle to enable the matching process.

NZTA has addressed privacy in several ways:

• Data deletion: Images of non-offending vehicles are deleted within minutes at the camera site
• Offence retention: Images of offending vehicles are retained for three years
• Facial blurring: The faces of drivers and passengers are automatically blurred in captured images
• Legal framework: Data is managed under the Privacy Act 2020
• Data sharing: NZTA states it only shares camera data to process infringements or as required by law
• Research use: When camera data is used for research, it is anonymised

But concerns remain. Privacy consultants engaged by NZTA in 2022 identified four "red"-rated risks to personal data. These were risks deemed "almost certain to occur" with "severe" impacts. They also found 11 additional high-risk vulnerabilities. The consultants noted that ANPR technology tends to be controversial and warned about the potential for building a database of vehicle movements over time.

The camera infrastructure involves cloud services provided by Amazon and Microsoft, with data temporarily stored on Australian servers until a New Zealand data centre is constructed. That's raised data sovereignty questions, and rightly so.

Common Misconceptions#

"I can speed between cameras and slow down at the end"#

Dangerous and unlikely to work consistently. The entire point of average speed enforcement is that your behaviour across the whole corridor matters. Speeding at 140 km/h for half the corridor and crawling at 60 km/h for the other half may produce a compliant average, but the risk of a fatal crash during the high-speed portion is extreme.

"They only work in one direction"#

Average speed cameras typically monitor both directions of travel with separate camera pairs for each direction.

"Changing lanes defeats the camera"#

Modern ANPR cameras capture plates across all lanes. Lane position is irrelevant to the system.

"They cannot work at night or in bad weather"#

The infrared ANPR technology is specifically designed for all-conditions operation. Night-time and poor weather have minimal impact on plate recognition accuracy.

Tips for Drivers#

1. Drive at or below the posted limit: Simplest and safest approach. If you drive at the limit, the system will never flag you.
2. Use cruise control: On corridors where it's safe to do so, cruise control helps maintain a consistent speed.
3. Don't try to game the system: Speeding and then slowing down is dangerous and unnecessary if you simply drive at the limit.
4. Watch for signs: Average speed camera zones are signposted with "Safety Camera Area" signs at the entry point.
5. Remember the limit is the maximum: The posted speed limit is the legal maximum, not a target. Driving slightly below it gives you a comfortable margin.

Related Information#

• How All Speed Cameras Work
• All Camera Locations
• About Our Data

Sources#

• NZTA - Average Speed Safety Cameras
• NZTA - New Average Speed Safety Camera Sites
• NZTA - Average Speed Cameras Factsheet (PDF)
• NZTA - Safety Camera Privacy
• NZTA - Average Speed Safety Cameras to Be Rolled Out
• RNZ - NZTA Defends Oversight of Highway Speed Camera System
• RAC Foundation - Effectiveness of Average Speed Cameras in Great Britain
• Jenoptik - Average Speed Cameras for Section Speed Control
• 1News - Average Speed Safety Cameras Will Lead to Fewer Tickets
• 1News - Locations of Six New Average Speed Safety Cameras Confirmed

---

This content has been fact-checked against official sources but may contain inaccuracies. This is general information only and does not constitute legal advice. Data updated weekly from Waka Kotahi NZTA.