What automatic number plate recognition does

Automatic number plate recognition (ANPR) is software that reads vehicle license plates from images or video feeds and converts the plate number into searchable data. The system uses a camera to capture the plate, optical character recognition (OCR) technology to read the characters, and a database to store or match that information. It works in real time or on recorded footage, and the speed depends on camera quality and lighting conditions.

ANPR is used by law enforcement to locate stolen vehicles, identify vehicles with outstanding warrants, or track vehicles involved in crimes. Parking facilities use it to manage paid parking and enforce time limits. Toll roads use it to bill drivers automatically. Border agencies use it to monitor vehicle movement across checkpoints. Private companies use it for fleet management and security at gated facilities.

The software does not make decisions on its own. A human operator reviews the matches the system flags, verifies them against records, and decides what action to take. The system is a tool that narrows the search space — it cannot arrest someone or issue a ticket without human review and authorization.

Key Takeaways

  • ANPR software reads license plates from cameras and converts them to searchable text using optical character recognition technology.
  • Law enforcement, parking operators, toll agencies, and private security all use ANPR for different purposes, from locating stolen vehicles to managing parking lots.
  • The system flags potential matches against databases, but a human operator must review and verify each match before any action is taken.
  • ANPR accuracy depends on camera quality, lighting, plate condition, and the angle at which the camera captures the plate.
  • Different jurisdictions have different rules about how long ANPR data can be stored and who is allowed to access it.

How the software reads a license plate

The process starts with a camera — usually a high-speed, high-resolution camera mounted on a pole, vehicle, or building. The camera captures an image of a passing vehicle's plate. The image is sent to the ANPR software, which isolates the plate area from the rest of the image using edge detection and contrast analysis.

Once the plate region is isolated, the software applies optical character recognition (OCR) to read each character. OCR compares the shapes it sees against a library of known letter and number shapes, accounting for different fonts, angles, and wear. The software outputs a text string — for example, "ABC1234" — along with a confidence score that indicates how certain the system is about each character.

If the confidence score is below a set threshold, the software may flag the plate as unreadable or uncertain. Operators can then review the image manually. The plate number is then checked against one or more databases — a stolen vehicle registry, a wanted persons list, a parking violation database, or a toll account database, depending on what the system is being used for.

Where ANPR is deployed and why

Law enforcement agencies use ANPR on patrol vehicles, at fixed checkpoints, and on highway cameras to search for stolen vehicles, vehicles with suspended registrations, or vehicles connected to criminal investigations. The system can scan hundreds of plates per minute on a busy highway.

Parking operators use ANPR to enforce paid parking zones and time-limited spaces. A camera reads plates as vehicles enter and leave, and the system automatically flags vehicles that have overstayed their time limit or have not paid. This eliminates the need for parking enforcement officers to manually check each vehicle.

Toll road operators use ANPR to bill drivers automatically. The system reads the plate, looks up the vehicle owner's account, and charges the toll. Drivers who do not pay receive a bill or citation in the mail.

Private security uses ANPR at gated facilities, corporate campuses, and parking garages to track who is entering and leaving, to alert security if a vehicle on a watchlist is detected, or to enforce parking rules within a facility.

Accuracy and limitations of ANPR systems

ANPR accuracy varies widely depending on several factors. High-quality cameras in good lighting conditions can achieve read rates above 95 percent. Poor lighting, rain, snow, mud on the plate, or a steep camera angle can drop accuracy to 70 percent or lower. Plates that are bent, faded, or obscured are harder to read. Vanity plates with unusual fonts or non-standard characters can confuse the OCR engine.

The software can also misread similar-looking characters — the letter O and the number 0, or the letter I and the number 1. If the OCR engine is not well-trained on the specific plate format used in a jurisdiction, errors increase. Some systems are better at reading certain plate styles than others.

False positives are a real problem. A plate that is misread as matching a wanted vehicle will flag an innocent driver. This is why human review is essential — an operator must visually confirm the match before any enforcement action is taken. In high-volume systems, false positives can create significant workload for operators.

Data storage and privacy considerations

ANPR systems generate large volumes of data. Every plate read is recorded with a timestamp and location. Over time, this creates a detailed record of where vehicles have been and when. Different jurisdictions have different rules about how long this data can be kept and who can access it.

Some jurisdictions require that ANPR data be deleted after a set period — 30 days, 90 days, or one year — unless the plate matched a wanted vehicle or active investigation. Other jurisdictions allow longer retention. Some require that access to ANPR data be logged and audited. Others have fewer restrictions.

Law enforcement agencies in the United States are not required by federal law to disclose ANPR data to the public, and many treat it as investigative information. Some states and cities have passed local laws requiring transparency reports or limiting how long data can be stored. The rules vary significantly by location.

Private companies that operate ANPR systems — parking operators, toll agencies, security firms — typically keep data longer than law enforcement, because they use it for billing and account management. This data is usually not subject to public records requests.

Common ANPR software and systems

Several commercial ANPR platforms are widely used. Vigilant Solutions (now part of Motorola Solutions) operates one of the largest ANPR databases in North America, used by thousands of law enforcement agencies. Genetec, Axis Communications, and Hikvision make ANPR software that integrates with security camera systems. Parkwhiz and ParkMobile use ANPR for parking management. Toll agencies often use custom systems built by vendors like Kapsch or Conduent.

Open-source ANPR software also exists — OpenALPR is a free, open-source platform that anyone can read and deploy. It is less accurate than commercial systems but is used by some smaller agencies and private operators who want to avoid licensing costs.

The choice of system depends on the use case. Law enforcement prioritizes accuracy and integration with criminal databases. Parking operators prioritize speed and integration with payment systems. Toll agencies prioritize reliability and uptime.

How ANPR data flows to decision-makers

When an ANPR system flags a plate match, the alert goes to an operator or dispatcher. In law enforcement, this might be a police dispatcher who receives an alert that a stolen vehicle has been detected. The dispatcher verifies the match by looking at the image, checks the vehicle details against the database record, and then decides whether to send an officer to investigate.

In parking enforcement, the system automatically generates a citation and sends it to the vehicle owner's address based on the registration information. The owner can contest the citation if they believe it was issued in error.

In toll collection, the system automatically bills the account associated with the plate. If the account is not valid or the bill is not paid, a notice is sent to the vehicle owner.

The key point is that ANPR is a detection and flagging tool, not an enforcement tool. A human must review the match, verify it is correct, and authorize the next step. The speed of that review varies — law enforcement might respond in minutes, while a parking citation might take days to generate and mail.

Frequently Asked Questions

Can ANPR read plates from any angle or distance?

No. ANPR works best when the camera is positioned to capture the plate head-on or at a shallow angle, typically within 30 feet. Extreme angles, very long distances, or obstructed plates reduce accuracy significantly. This is why law enforcement mounts ANPR cameras at specific heights and angles on highways and at checkpoints.

Does ANPR work at night or in bad weather?

Modern ANPR systems can read plates at night using infrared or high-intensity lighting, though accuracy may be slightly lower than in daylight. Heavy rain, snow, or fog reduces accuracy because it obscures the plate or reduces image contrast. Mud or dirt on the plate also causes problems.

How long does it take ANPR to read a plate?

The actual reading process takes less than a second. A high-speed camera can capture and process hundreds of plates per minute on a busy highway. However, the time from detection to human review and action varies — it could be seconds for law enforcement responding to a stolen vehicle alert, or days for a parking citation to be generated and mailed.

Can someone block or spoof an ANPR camera?

Deliberately obscuring, covering, or altering a license plate to prevent ANPR from reading it is illegal in most jurisdictions. Reflective coatings, special frames, or photoshopped images in online databases do not work against modern ANPR systems. Spoofing — creating a fake plate — is also illegal and is a separate crime from the original offense.

Who owns the ANPR data collected by law enforcement?

This varies by jurisdiction. In some places, ANPR data is considered public record and can be requested under freedom of information laws. In others, it is treated as investigative information and is not disclosed. Some jurisdictions have passed laws requiring that ANPR data be deleted after a certain period unless it is connected to an active investigation. Check your local or state laws for the rules in your area.