How ANPR Cameras Really Work: From Pixel to Prosecution

Executive Summary

ANPR (Automatic Number Plate Recognition) cameras use infrared illumination, optical character recognition (OCR), and database matching to automatically read vehicle registration numbers at speeds of over 100 mph, processing thousands of vehicles daily with 90 to 95% accuracy. The technology captures multiple images per second, isolates the number plate region, converts characters to digital text, and instantly checks against databases containing millions of records, all within 300 to 500 milliseconds.

For UK drivers, ANPR readability depends entirely on having BS AU 145e:2018 compliant number plates with correct reflectivity, the proper Charles Wright font, and accurate character spacing. Non-compliant plates, including those with incorrect fonts, poor reflectivity, or illegal modifications, fail ANPR detection. This results in missed enforcement, potential police stops, and fines of up to £1,000 under the Road Vehicles (Display of Registration Marks) Regulations 2001.

As a DVLA-registered manufacturer (RNPS ID: 73132), we produce every plate using BS AU 145e certified materials specifically tested for ANPR compatibility, including near-infrared (NIR) reflectivity that allows cameras to read plates in total darkness, rain, and at high speeds. This guide explains the complete ANPR process from image capture to enforcement, why plate compliance matters, and how to ensure your vehicle remains readable by the UK’s 10,000+ ANPR cameras.

What is ANPR? Understanding the Technology

Plain English Overview

ANPR (Automatic Number Plate Recognition), also called ALPR (Automatic License Plate Recognition) in some countries, is a surveillance technology that automatically reads vehicle registration numbers without human intervention. The system combines specialised cameras, infrared lighting, and computer software to:

• Detect vehicles entering the camera’s field of view
• Capture high-speed images of the number plate
• Extract the registration number as text
• Check against databases (stolen vehicles, untaxed, no MOT, etc.)
• Alert authorities or record the sighting

The entire process takes less than half a second, faster than a human blink.

Scale of ANPR in the UK

The UK operates one of the world’s largest ANPR networks:

• 10,000+ fixed and mobile ANPR cameras across the UK (National ANPR Service)
• 100+ million reads per day on average
• The National ANPR Data Centre processes all readings
• Integration with police forces, DVLA, DVSA, and local authorities
• 24/7 operation regardless of weather or lighting conditions

For a full breakdown of the penalties involved, see our guide on number plate fines in the UK.

Historical Development

ANPR technology evolved through several generations:

• 1970s to 1980s: Early development by the UK Police Scientific Development Branch
• 1990s: First operational deployments for stolen vehicle detection
• 2000s: Integration with road tax and MOT databases
• 2010s: Near-infrared (NIR) compatibility requirements introduced
• 2018: BS AU 145e:2018 standard mandated enhanced ANPR readability
• 2020s: AI-powered recognition with 95%+ accuracy rates

The Technology: How ANPR Actually Works

Step 1: Image Capture

Camera Specifications:

ANPR cameras differ significantly from standard CCTV:

• Shutter speed: 1/1000 to 1/10000 second (freezes motion at over 100 mph)
• Frame rate: 25 to 60 frames per second
• Resolution: 1 to 5 megapixels (optimised for the plate region, not the overall image)
• Dynamic range: High dynamic range (HDR) to handle varying lighting
• Wavelength sensitivity: Visible light (400 to 700nm) plus near-infrared (700 to 1000nm)

Positioning:

Cameras are strategically placed to optimise capture:

• Fixed gantries: Motorways, major roads (permanent installation)
• Mobile units: Police vehicles, temporary enforcement zones
• Entry/exit points: Car parks, congestion charge zones, toll booths
• Angle: Typically 0 to 30 degrees from perpendicular to the plate
• Height: 2 to 5 metres above the road surface

Step 2: Infrared Illumination

Why Infrared Matters:

ANPR cameras use near-infrared (NIR) light for several critical reasons.

1. Consistent Illumination:

• Works in total darkness without visible light pollution
• Unaffected by headlights, streetlights, or weather conditions
• Provides uniform lighting across the plate surface

2. Reflective Plate Response:

• BS AU 145e plates use retroreflective sheeting designed for NIR wavelengths
• White/yellow backgrounds reflect NIR light strongly (they appear bright)
• Black characters absorb NIR light (they appear dark)
• Creates maximum contrast for character recognition

3. Wavelength Specificity:

• Typical ANPR systems use a 730 to 950nm wavelength
• This range optimises plate reflectivity while minimising environmental interference
• Invisible to the human eye (no distracting flashes)

Technical Specification:

• Infrared LEDs emit light at specific wavelengths (commonly 850nm or 940nm)
• Pulsed illumination synchronised with the camera shutter
• Intensity: 10 to 100 watts depending on range requirements
• Beam pattern focused on the plate region only

Step 3: Plate Detection and Isolation

Finding the Plate:

Before reading characters, the system must locate the plate within the image.

Algorithm Process:

Edge Detection:

• Software identifies rectangular shapes matching plate dimensions
• UK plates: 520mm x 111mm (front/rear standard size)
• Aspect ratio analysis filters out non-plate objects

Contrast Analysis:

• Searches for high-contrast regions (dark characters on a light background)
• The reflective background creates a distinctive brightness signature
• Eliminates false positives (grilles, badges, stickers)

Geometric Verification:

• Checks for correct character spacing (11mm for cars)
• Verifies character height (79mm for cars, 64mm for motorcycles)
• Confirms plate position relative to the vehicle

Region of Interest (ROI) Extraction:

• Crops the image to the plate area only
• Reduces the processing load for the OCR stage
• Typical ROI: 200×80 pixels from the full image

Processing time: 50 to 100 milliseconds.

Step 4: Character Segmentation

Isolating Individual Characters:

Once the plate region is identified, each character must be separated.

Segmentation Techniques:

Thresholding:

• Converts the greyscale image to pure black and white
• Pixels above the brightness threshold = white (background)
• Pixels below the threshold = black (characters)
• Adaptive thresholding handles varying lighting

Morphological Operations:

• Removes noise (dirt, scratches, reflections)
• Fills gaps in characters
• Separates touching characters

Character Boundary Detection:

• Identifies individual character boundaries
• Measures spacing between characters (should be 11mm)
• Handles multi-line plates (motorcycles, imported vehicles)

Normalization:

• Scales all characters to a standard size
• Corrects for perspective distortion
• Adjusts for camera angle variations

Challenges:

• Dirty or obscured plates
• Damaged characters
• Non-standard fonts (illegal plates)
• Extreme camera angles
• Motion blur at high speeds

Processing time: 30 to 50 milliseconds.

Step 5: Optical Character Recognition (OCR)

Converting Images to Text:

OCR is the core technology that translates character images into alphanumeric text.

Modern ANPR OCR Methods:

1. Feature Extraction:

• Analyses character shape characteristics
• Identifies curves, lines, angles, and intersections
• Creates a mathematical representation of each character
• Example: “A” has 3 lines meeting at specific angles

2. Pattern Matching:

• Compares extracted features against a character database
• Contains thousands of variations for each character (0 to 9, A to Z)
• Accounts for font variations, damage, and partial occlusion
• Uses statistical models for probability matching

3. Neural Networks (Modern Systems):

• Deep learning convolutional neural networks (CNNs)
• Trained on millions of plate images
• Self-learning improves accuracy over time
• Handles edge cases better than traditional methods
• Accuracy: 95 to 98% in optimal conditions

4. Contextual Validation:

• Checks if the result matches UK registration formats
• Validates against known patterns (e.g., AB51 CDE)
• Rejects impossible combinations
• Cross-references with a dictionary of valid registrations

UK Registration Format Checking:

The system validates against DVLA registration formats:

• Current style (2001 to present): AA## AAA (e.g., AB51 CDE)
• Prefix style (1983 to 2001): A### AAA (e.g., A123 BCD)
• Suffix style (1963 to 1983): AAA ###A (e.g., ABC 123D)
• Dateless: Up to 3 letters plus up to 3 numbers (e.g., ABC 123)
• Northern Ireland: Same formats but includes I and Z characters

Processing time: 100 to 200 milliseconds.

Step 6: Database Matching and Action

What Happens After Reading:

Once the registration is extracted, the system performs instant database checks.

Real-Time Checks:

Police National Computer (PNC):

• Stolen vehicles
• Vehicles linked to serious crimes
• AMBER alerts (child abduction)
• Wanted persons associated with the vehicle

DVLA Database:

• Vehicle tax status (untaxed vehicles)
• MOT status (no valid MOT)
• Insurance status (via the Motor Insurance Database)
• Registered keeper details

Local Databases:

• Congestion charge payment (London)
• ULEZ compliance (London, Birmingham, etc.)
• Parking permits
• Toll payment

Watchlists:

• Vehicles of interest to specific investigations
• Known offender vehicles
• Intelligence-led policing targets

Response Actions:

Immediate Alert (Real-Time):

• Police vehicle-mounted ANPR: Audible/visual alert to the officer
• Fixed camera with police link: Dispatch to the nearest unit
• Priority: Stolen vehicle, wanted person, serious crime link

Automated Enforcement:

• Untaxed vehicle: Automatic £100 fine by post
• No MOT: £100 fixed penalty
• No insurance: £200 fixed penalty plus 6 to 8 points
• Congestion charge violation: £160 penalty (reduced to £80 if paid within 14 days)

Data Logging:

• All reads stored in the National ANPR Data Centre
• Retention period: Typically 1 to 2 years (varies by purpose)
• Used for pattern analysis, investigations, and intelligence

Processing time: 50 to 150 milliseconds.
Total system time: 300 to 500 milliseconds from vehicle entry to action.

The Journey: From Camera to Prosecution

Real-World Example 1: Speed Enforcement

Scenario: A vehicle exceeds the speed limit on an A-road with average speed cameras.

Timeline:

00:00:00 – Vehicle enters ANPR zone at Camera A

• Camera captures plate: AB51 CDE
• Timestamp: 14:23:15.000
• GPS location: Mile marker 23.5

00:00:01 – ANPR software processes the image

• Plate read: AB51 CDE (98% confidence)
• Vehicle class: Car (based on plate size/position)
• Data transmitted to the central server

00:00:02 – Database check

• Registration valid: Yes
• Vehicle tax: Current
• MOT: Valid
• Insurance: Confirmed
• No alerts, routine read logged

00:15:00 – Same vehicle reaches Camera B (10 miles away)

• Timestamp: 14:38:15.000
• Travel time: 15 minutes
• Distance: 10 miles
• Average speed: 40 mph
• Speed limit: 30 mph
• Violation detected

00:15:01 – Automated enforcement triggered

• System calculates: 40 mph in a 30 mph zone (33% over the limit)
• Threshold exceeded (typically 10% plus 2 mph = 35 mph threshold)
• Case created in the enforcement system

00:15:02 – Evidence package compiled

• Image from Camera A (entry)
• Image from Camera B (exit)
• Timestamps and locations
• Calculated speed
• Registration details from DVLA

00:15:05 – Human verification

• Trained operator reviews the images
• Confirms plate readability
• Verifies vehicle identity
• Approves enforcement action

Day +14 – Notice of Intended Prosecution (NIP) issued

• Sent to the registered keeper’s address
• Includes: Images, speed, location, date/time
• Requires: Driver identification within 28 days

Day +28 – Response received

• Keeper identifies the driver
• Fixed Penalty Notice issued: £100 plus 3 points
• Or a court summons if contested

Total processing time: 2 to 4 weeks from violation to NIP.

Real-World Example 2: Stolen Vehicle Detection

Scenario: A stolen vehicle passes a police ANPR camera.

Timeline:

00:00:00 – Vehicle reported stolen

• Owner calls 999
• Registration: XY12 ABC
• Vehicle: Black BMW 3 Series
• Stolen from: Manchester, M1 1AA
• Time: 02:15

00:00:05 – Vehicle entered on PNC

• Police National Computer updated
• Flag: STOLEN VEHICLE
• Priority: HIGH
• Alert to all ANPR systems nationwide

00:45:00 – Vehicle passes ANPR camera in Birmingham

• Distance from theft: 85 miles
• Camera: Fixed gantry on the M6
• Time: 03:00

00:45:00.300 – ANPR reads the plate

• Registration: XY12 ABC
• Confidence: 96%
• Immediate PNC check initiated

00:45:00.500 – Match found

• Status: STOLEN
• Alert generated instantly

00:45:01 – Police alert

• Nearest patrol vehicle notified
• Location: M6 southbound, junction 6
• Vehicle description: Black BMW 3 Series
• Warning: Stolen, high risk

00:45:02 – Officer response

• Officer acknowledges the alert
• Positions for intercept
• Requests backup

00:47:00 – Vehicle intercepted

• Safe stop performed
• Driver arrested
• Vehicle recovered

Total time from read to arrest: 2 minutes.

Real-World Example 3: ULEZ/Congestion Charge Enforcement

Scenario: A non-compliant vehicle enters the London ULEZ zone.

Timeline:

00:00:00 – Vehicle enters ULEZ boundary

• ANPR camera at the entry point
• Registration: AB12 CDE
• Vehicle: 2015 Diesel Ford Focus
• Euro standard: Euro 5 (non-compliant)

00:00:00.400 – Plate read and database check

• DVLA lookup: 2015 Ford Focus Diesel
• Emissions database: Euro 5
• ULEZ standard: Euro 6 required for diesel
• Non-compliant identified

00:00:01 – Payment database check

• Checks if the daily charge (£12.50) was paid
• Payment status: NOT PAID
• Violation recorded

00:00:02 – Data logged

• Entry time: 08:15:23
• Location: ULEZ boundary point
• Vehicle details stored
• Evidence image saved

Day +1 – Automated processing

• System compiles violation data
• Registered keeper details from DVLA
• PCN (Penalty Charge Notice) generated

Day +3 – PCN issued by post

• Sent to the keeper’s address
• Penalty: £160 (reduced to £80 if paid within 14 days)
• Includes: Images, date/time, location, vehicle details
• Payment instructions and appeal process

Day +28 – If unpaid

• Penalty increases to £160
• Enforcement action begins
• Possible: Debt collection, court action

Why BS AU 145e Compliance Matters for ANPR

The Reflectivity Requirement

BS AU 145e:2018 Standard:

The British Standard specifies exact reflectivity requirements that directly impact ANPR performance.

Luminance Coefficient (Reflectivity):

• White background: Minimum 300 to 400 cd/lx/m²
• Yellow background: Minimum 200 to 300 cd/lx/m²
• Black characters: Maximum 10 to 15 cd/lx/m² (must absorb light)
• Contrast ratio: Minimum 20:1 between background and characters

Why This Matters:

1. Infrared Response:

• Reflective sheeting must reflect NIR light (700 to 1000nm wavelength)
• Non-compliant plates may reflect visible light but fail in the NIR spectrum
• ANPR cameras primarily use NIR illumination
• Poor NIR reflectivity = unreadable plates

2. Consistency:

• BS AU 145e ensures all legal plates have similar reflectivity
• ANPR systems are calibrated for a standard reflectivity range
• Too dim: Plate invisible to the camera
• Too bright: Character detail lost in glare

3. Weather Resistance:

Reflectivity must be maintained after:

• 2+ years of UV exposure
• Salt spray (coastal/winter road conditions)
• Temperature cycling (-20°C to +50°C)
• High-pressure washing

Non-compliant plates fade, delaminate, or lose reflectivity. Our guide on what makes a number plate legal under BS AU 145e covers the detailed standard requirements.

Character Spacing and Font Requirements

Charles Wright Font:

Mandatory since 2001 under the Road Vehicles (Display of Registration Marks) Regulations 2001.

Technical Specifications:

• Character height: 79mm (cars), 64mm (motorcycles)
• Character width: 50mm (except ‘1’ and ‘I’, which are narrower)
• Stroke thickness: 14mm
• Character spacing: 11mm between characters
• Line spacing: 11mm (for two-line plates)
• Margins: 11mm minimum on all sides

Our guide on the Charles Wright font and why it’s mandatory explains these specifications in more detail.

Why Exact Spacing Matters:

1. Segmentation Accuracy:

• ANPR algorithms expect specific spacing patterns
• Incorrect spacing confuses character separation
• 11mm spacing creates predictable character boundaries
• Tighter spacing: Characters merge together
• Wider spacing: The system may read them as separate plates

2. Format Validation:

• The system checks if the read matches known registration formats
• Correct spacing ensures an accurate character count
• Prevents false reads (e.g., “AB 51CDE” vs “AB51 CDE”)

3. Perspective Correction:

• Software corrects for camera angle using known dimensions
• Standard spacing provides reference points
• Non-standard spacing reduces correction accuracy

Common Font Violations:

Illegal Fonts:

• “VIP” style (elongated characters)
• “Carbon” style (italicised)
• “Ghost” plates with gradient effects affecting contrast
• Custom fonts for “styling”

Impact on ANPR:

• Read failure rates increase from 5% to 40%+
• Misreads are common (e.g., “8” read as “B”, “0” as “O”)
• May trigger a police stop for plate inspection
• Results in fines of up to £1,000

If you’re curious about where ghost plates stand legally, our guide on whether ghost plates are legal explains the current rules and the risks.

Plate Position and Angle

Optimal ANPR Readability:

Vertical Angle:

• Best: 0 to 15 degrees from perpendicular
• Acceptable: 0 to 30 degrees
• Poor: 30+ degrees (significant accuracy loss)

Horizontal Angle:

• Best: Directly facing the camera
• Acceptable: Up to 30 degrees off-axis
• Poor: 45+ degrees (characters distorted)

Height:

• Standard mounting: 400 to 1000mm from the ground
• Too high: Camera may miss the plate
• Too low: Obscured by road debris, angle issues

Common Positioning Problems:

1. Towed Vehicles/Trailers:

• Plate may be at an extreme angle
• Reduced readability
• May require multiple camera passes

2. Modified Vehicles:

• Lowered suspension changes the plate angle
• Raised 4x4s may have plates outside the optimal range
• Aftermarket bumpers may obscure the plate

3. Damaged Mounting:

• Loose plates swing or tilt
• Creates motion blur
• Variable angle reduces consistency

Accuracy and Error Rates: The Reality

ANPR Success Rates

Overall Performance:

Modern ANPR systems achieve:

• 90 to 95% accuracy in optimal conditions
• 85 to 90% accuracy in challenging conditions
• 98%+ accuracy with a human verification layer

Factors Affecting Accuracy:

Environmental Conditions:

Condition	Accuracy	Reason
Clear daylight	95 to 98%	Optimal lighting, high contrast
Night (dry)	93 to 96%	IR illumination effective
Heavy rain	85 to 90%	Water droplets scatter IR light
Snow/ice	80 to 85%	Plate obscurement, reflection issues
Fog/mist	85 to 90%	Reduced IR range, scattering
Direct sunlight	90 to 93%	Glare, high dynamic range challenges

Vehicle Speed:

Speed	Accuracy	Notes
0 to 30 mph	96 to 98%	Stationary or slow = optimal
30 to 60 mph	93 to 96%	Standard motorway speeds
60 to 80 mph	90 to 94%	High speed, motion blur risk
80+ mph	85 to 90%	Extreme speed challenges

Plate Condition:

Condition	Accuracy	Issue
Clean, new plate	96 to 98%	Ideal condition
Minor dirt/dust	93 to 96%	Slight contrast reduction
Moderate dirt	85 to 92%	Character obscurement
Heavy dirt/snow	60 to 80%	May be unreadable
Faded/damaged	70 to 85%	Reduced reflectivity
Illegal modifications	40 to 70%	Unpredictable failures

Common ANPR Errors

False Reads (Misreads):

Character Confusion:

• “8” ↔ “B” (similar shape)
• “0” ↔ “O” (zero vs letter O)
• “1” ↔ “I” ↔ “l” (one, capital I, lowercase L)
• “2” ↔ “Z” (in some fonts)
• “5” ↔ “S” (similar curves)

Mitigation:

• Context checking against valid registration formats
• Dictionary lookup (is this a real registration?)
• Multiple camera passes for confirmation
• Human verification for enforcement cases

False Negatives (Missed Reads):

Causes:

• Plate completely obscured (mud, snow, damage)
• Extreme camera angle (>45 degrees)
• Plate outside the BS AU 145e standard
• System malfunction
• Overexposure or underexposure

Impact:

• Violation not detected
• Stolen vehicle not flagged
• Data gap in vehicle tracking
• Generally lower risk than false positives

False Positives (Incorrect Alerts):

Scenario: The system reads “AB51 CDE” but the actual plate is “AB5I CDE”.

Consequences:

• Innocent driver stopped by police
• Wrong person fined
• Database match to the wrong vehicle
• Serious privacy and trust issues

Prevention:

• Confidence thresholds (only alert if >90% confidence)
• Human verification before enforcement action
• Multiple reads required for serious alerts
• Appeal processes for erroneous penalties

Human Verification Layer

Quality Assurance:

Despite high accuracy, critical decisions involve human review.

When Human Verification is Required:

Speed Enforcement:

• All average speed camera violations
• Operator confirms plate readability
• Verifies vehicle identity
• Approves NIP issuance

Police ANPR Alerts:

• Officer reviews the image before action
• Confirms the match to alert criteria
• Assesses the risk level
• Decides on intervention

Low Confidence Reads:

• System flags reads below the confidence threshold
• Manual review by a trained operator
• May request additional evidence

Verification Process:

Trained Operators:

• Specialist ANPR training (typically 2 to 4 weeks)
• Certification required
• Regular competency assessments
• Access to enhanced imaging tools

Tools Available:

• Image enhancement (contrast, brightness, zoom)
• Multiple frame review (video sequence)
• Historical reads for the same vehicle
• Cross-reference with other cameras

Decision Criteria:

• Plate clearly visible and readable
• Characters unambiguous
• Matches the database record
• No reasonable doubt

Accuracy with human review: 99%+.

Privacy and Data Protection

Legal Framework

Governing Legislation:

ANPR operation in the UK is regulated by several pieces of legislation.

1. Data Protection Act 2018 / UK GDPR:

• ANPR data = personal data (it links to the vehicle keeper)
• Requires a lawful basis for processing
• Data minimisation principles
• Retention limits
• Individual rights (access, erasure)

2. Protection of Freedoms Act 2012:

• Regulates biometric and surveillance data
• ANPR Code of Practice requirements
• Oversight mechanisms

3. Regulation of Investigatory Powers Act 2000 (RIPA):

• Governs directed surveillance
• ANPR for criminal investigations
• Authorisation requirements

4. ANPR Code of Practice:

• Published by the Surveillance Camera Commissioner
• Mandatory for police and government use
• Best practice for private operators
• Regular updates and compliance checks

Data Retention Periods

Standard Retention:

National ANPR Data Centre:

• Routine reads: 1 to 2 years
• Linked to investigations: Up to 6 years
• Serious crime/terrorism: Indefinite (reviewed regularly)

Local Police Forces:

• Operational data: 90 days to 2 years
• Intelligence purposes: As required
• Case evidence: Until the case is concluded plus appeals

Private Operators (Parking, Toll Roads):

• Typically 12 to 24 months
• Varies by purpose and operator
• Must comply with UK GDPR

Your Rights:

Subject Access Request (SAR):

• Right to request your ANPR data
• Must be provided within 30 days
• Free of charge (unless requests are excessive)
• Includes: Reads, timestamps, locations, images

How to Request:

• Contact the data controller (police force, DVLA, operator)
• Provide: Your registration, date range, purpose
• Proof of identity required
• Specify the format (digital, paper)

Limitations:

• Ongoing investigations may be exempt
• Third-party data redacted
• National security exemptions
• Disproportionate effort exemptions

Data Security

Protection Measures:

Encryption:

• Data encrypted in transit (TLS/SSL)
• Encryption at rest (AES-256)
• Secure key management

Access Controls:

• Role-based access only
• Multi-factor authentication
• Audit trails (who accessed what, and when)
• Regular access reviews

Physical Security:

• Data centres with 24/7 security
• Biometric access controls
• CCTV monitoring
• Disaster recovery sites

Compliance Audits:

• Regular ICO (Information Commissioner’s Office) audits
• Surveillance Camera Commissioner oversight
• Internal compliance reviews
• Breach notification requirements (72 hours)

Practical Takeaways

Ensuring Your Plate is ANPR-Readable

✅ DO:

Use BS AU 145e:2018 compliant plates:

• Verify the supplier is DVLA-registered (RNPS ID displayed)
• Check plates have the correct reflectivity
• Ensure NIR compatibility for night reading

Maintain plate condition:

• Clean regularly (weekly in winter, monthly otherwise)
• Remove dirt, mud, and snow immediately
• Check for fading or damage
• Replace if reflectivity deteriorates

Verify correct fitting:

• Charles Wright font mandatory
• Correct character spacing (11mm)
• Proper mounting angle (perpendicular to the ground)
• Secure fixing (no swinging or movement)

Check legality:

• No tinted covers (illegal and fails ANPR)
• No bolts obscuring characters
• No dirt or obstruction
• Both plates visible and readable

❌ DON’T:

Use illegal plate styles:

• “Ghost” plates with poor contrast
• Custom fonts (VIP, carbon, italic)
• 3D/4D plates that fail spacing requirements
• Reflective sprays or films

Ignore damage:

• Cracked or broken plates
• Faded characters
• Peeling reflective sheeting
• Loose mounting

Assume “it won’t happen to you”:

• ANPR reads 100+ million vehicles daily
• Non-compliant plates WILL be detected
• Fines of up to £1,000 for illegal plates
• MOT failure guaranteed

For a fuller picture of what makes a plate non-compliant, see our guide on what makes a number plate illegal.

What to Do If ANPR Makes an Error

If You Receive an Incorrect Penalty:

Don’t Ignore It:

• Penalties increase if unpaid
• May lead to court action
• Could affect your credit rating

Gather Evidence:

• Photograph your plate (showing it’s legal and clean)
• Note the plate condition on the date in question
• Check the weather conditions
• Obtain witness statements if available

Request Evidence:

• Ask for the ANPR images
• Request camera calibration records
• Check if human verification occurred
• Review timestamp accuracy

Formal Appeal:

• Follow the appeal process on the penalty notice
• Submit evidence clearly
• Keep copies of everything
• Meet all deadlines

Escalate if Necessary:

• Independent adjudicator (for parking/Congestion Charge)
• Court (for criminal matters)
• ICO complaint (for data protection issues)

If Stopped by Police Due to an ANPR Error:

1. Remain calm and cooperative
2. Explain the situation politely
3. Provide documentation (V5C, insurance, MOT)
4. Note the officer’s name and badge number
5. Request clarification of the alert
6. Follow up with a formal complaint if warranted

How Plate-Maker Ensures ANPR Compliance

Our Manufacturing Standards:

As a DVLA-registered manufacturer (RNPS ID: 73132), we guarantee the following.

BS AU 145e:2018 Certification:

• All plates use certified reflective sheeting
• Tested for NIR compatibility
• Reflectivity exceeds minimum requirements
• Weather-resistant for 5+ years

Precise Character Spacing:

• Computer-controlled cutting ensures 11mm spacing
• Charles Wright font exactly to specification
• Regular calibration of equipment
• Quality control checks on every plate

ANPR Testing:

• Sample plates tested with ANPR equipment
• Readability verified in various conditions
• Continuous improvement based on feedback
• 99%+ first-pass read rate

Document Verification:

• Full compliance with Schedule 2 requirements
• Prevents cloning and fraud
• Protects you from liability
• Maintains ANPR database integrity

Visit our same-day collection service in Ilford for compliant plates that pass ANPR checks.

Frequently Asked Questions

What percentage of number plates can ANPR cameras actually read?

Modern ANPR cameras achieve 90 to 95% accuracy in optimal conditions (clear weather, clean plates, correct speed). With BS AU 145e compliant plates in good condition, read rates exceed 96%. Accuracy drops to 80 to 85% in heavy rain, snow, or with dirty or damaged plates. Human verification increases final accuracy to 99%+ for enforcement cases.

Do ANPR cameras work at night or in bad weather?

Yes, ANPR cameras work 24/7 using near-infrared (NIR) illumination that is invisible to humans. Infrared light penetrates darkness, rain, and fog better than visible light. BS AU 145e plates are specifically designed to reflect NIR wavelengths, creating high-contrast images even in total darkness. Night accuracy (93 to 96%) is only slightly lower than daytime (95 to 98%).

Can I be fined if ANPR misreads my number plate?

Yes, but you can appeal. Initial penalties are automated, but serious enforcement requires human verification. If ANPR misreads your plate (e.g., “8” as “B”), you may receive an incorrect fine. You must appeal with evidence: photos of your legal plate, your V5C registration, and an explanation. Keep plates clean and BS AU 145e compliant to prevent misreads.

How long does ANPR data stay on record?

Routine ANPR reads are retained for 1 to 2 years in the National ANPR Data Centre. Data linked to criminal investigations may be kept for up to 6 years. Serious crime or terrorism cases can be retained indefinitely with regular review. You can request your ANPR data through a Subject Access Request (SAR) under UK GDPR, which must be fulfilled within 30 days.

What makes a number plate unreadable by ANPR cameras?

Common causes include non-BS AU 145e materials with poor reflectivity, dirty or obscured plates (mud, snow, grime), illegal fonts or incorrect character spacing, cracked or faded plates, tinted covers or reflective sprays, and extreme mounting angles. These can reduce ANPR accuracy from 95% to between 40 and 70%, triggering police stops and fines of up to £1,000 under the Road Vehicles Regulations 2001.

Are 3D or 4D number plates ANPR-compatible?

Yes, 3D gel and 4D laser-cut plates are ANPR-compatible when manufactured correctly. They must use BS AU 145e certified reflective backgrounds, maintain exact Charles Wright font spacing (11mm between characters), and have characters that don’t obscure the reflective surface. Our 3D gel plates and 4D laser-cut plates are tested for ANPR readability and display RNPS ID 73132 for compliance.

Conclusion

ANPR technology represents one of the most sophisticated automated surveillance systems in the UK, processing over 100 million vehicle reads daily with 90 to 95% accuracy. From the initial infrared illumination to database matching and enforcement, the entire process occurs in 300 to 500 milliseconds, faster than human reaction time.

For vehicle owners, the critical factor is ensuring your number plates meet BS AU 145e:2018 standards. The reflective sheeting, precise Charles Wright font, and correct character spacing are not arbitrary requirements. They are engineered specifications that enable ANPR cameras to read your plate reliably in all conditions, from bright sunlight to total darkness, and from stationary to over 100 mph.

Non-compliant plates, whether they have illegal fonts, poor reflectivity, or are dirty or obscured, fail ANPR detection. This results in missed reads, police stops, potential fines of up to £1,000, and MOT failure. The technology is unforgiving but fair: compliant plates read successfully, while non-compliant plates create problems.

At Private Number Plate Maker Ltd, we manufacture every plate at our Ilford workshop using BS AU 145e:2018 certified materials specifically tested for ANPR compatibility. As a DVLA-registered manufacturer (RNPS ID: 73132), we ensure your plates will be readable by the UK’s 10,000+ ANPR cameras, protecting you from enforcement issues while maintaining full legal compliance.

You can design your ANPR-compliant plates online, or visit our Eastern Avenue workshop for same-day collection. We’re here to ensure your vehicle meets all legal requirements and performs flawlessly with the UK’s ANPR network.