Remote Roadside ANPR System
Remote Roadside ANPR, Vehicle & Person Detection Prototype Plan
Section titled “Remote Roadside ANPR, Vehicle & Person Detection Prototype Plan”Project: Solar-Powered Remote Roadside ANPR, Vehicle & Person Detection Network Prototype
Location context: Residential neighbourhood in New Zealand
Document purpose: Practical prototype and implementation plan
Version: Final 3.0 — October 2026
Table of Contents
Section titled “Table of Contents”- Introduction
- Problem Statement
- Prototype Goal
- Recommended Fastest Prototype Strategy
- System Architecture
- Prototype Phases
- Camera Recommendation
- Night-Time Number Plate Capture
- Existing Software Options
- Recommended Open-Source Route — Frigate
- Alternative Open-Source / DIY Software
- Commercial / Ready-Made ANPR Software
- Custom Python ANPR Pipeline
- Person Detection, Face Capture and Optional Face Recognition
- Remote Solar Camera Node
- Power System Design
- Battery Sizing
- Solar Panel Sizing
- 4G / LTE Connectivity
- Weatherproof Enclosure
- Local vs Remote Processing
- Networking and Security
- Database Design
- Common, Rare and New Vehicle Detection
- Incident Search Workflow
- Storage and Retention
- Privacy and Deployment Considerations
- Prototype Test Plan
- Success Criteria
- Suggested Hardware List
- Suggested Software Stack
- Recommended Repository Structure
- Step-by-Step First Prototype
- Second Prototype — Remote Solar Node
- Future Multi-Camera Architecture
- Features to Avoid Initially
- Risks and Mitigations
- Final Recommended Build Order
- Definition of Done
- Reference Links
1. Introduction
Section titled “1. Introduction”Vehicle theft and suspicious vehicle activity have become a concern in the local residential area. A stolen vehicle may leave the area within minutes, while conventional household CCTV often records only a dark or blurred vehicle image at night.
The key problem is that ordinary wide-angle security cameras are normally designed to show the overall scene rather than capture a moving licence plate. At night, headlights, reflective number plates, motion blur and automatic camera exposure can make the plate unreadable even when the vehicle itself is visible.
This project proposes a small, privately operated ANPR prototype that can:
- detect vehicles passing a defined road location;
- capture readable number plates during the day and at night;
- record the plate, time, camera location and direction of travel;
- retain a supporting plate image and vehicle snapshot;
- provide fast searching around the time of a theft or other incident;
- distinguish frequently observed vehicles from rare or newly observed vehicles;
- eventually support several remote camera nodes;
- operate at locations where there is no mains power or fixed Internet connection.
Intended Roadside Deployment
Section titled “Intended Roadside Deployment”The intended physical deployment is a small standalone camera node positioned beside a road, at a carefully selected fixed location where it can observe passing vehicles.
The preferred prototype location is:
- on private property immediately beside or near the road;
- close enough to obtain a useful plate angle and plate size;
- positioned so that the camera looks at a narrow capture zone rather than the whole street;
- physically secured against weather and tampering.
The prototype is not limited to being mounted on a house. A later node may be completely independent of a nearby building and may operate using:
ANPR camera+ optional overview/person camera+ solar panel+ LiFePO4 battery+ MPPT charger+ 4G/LTE router+ weatherproof locked enclosureIf a proposed position is on a council road reserve, berm, street-light pole, utility pole, footpath area, or other land that is not privately owned, permission from the relevant land/infrastructure owner must be confirmed before installation.
The first objective is not to build a neighbourhood-wide surveillance system.
The first objective is to prove one thing:
Can one camera reliably capture and recognise moving New Zealand licence plates at night?
Only after that has been demonstrated should the project expand to remote solar nodes, multiple cameras, analytics or a custom dashboard.
2. Problem Statement
Section titled “2. Problem Statement”A normal CCTV camera may produce footage like:
Vehicle visible: YESVehicle colour: MAYBEDriver visible: NONumber plate readable: NOThe reasons commonly include:
- motion blur;
- long night-time exposure;
- headlight glare;
- plate overexposure;
- insufficient infrared illumination;
- plate occupying too few pixels;
- camera positioned at too large an angle;
- wide-angle lens;
- excessive digital zoom.
Therefore, this project must solve the optical capture problem first and the OCR problem second.
Recommended priority:
Camera / illumination ~70%Positioning / lens ~20%Software / OCR ~10%If the original frame does not contain readable characters, software cannot reliably reconstruct them.
3. Prototype Goal
Section titled “3. Prototype Goal”The complete prototype should eventually perform:
Vehicle passes road | vCamera captures plate | vANPR / OCR reads plate | vTimestamp + direction + confidence | vStore plate crop + vehicle image | vDatabase | vSearch by incident timeExample event:
Time: 2026-10-04 01:17:21Camera: NODE-01Plate: ABC123Confidence: 97%Direction: EastboundStatus: NEWPlate image: ABC123_011721_plate.jpgVehicle: ABC123_011721_vehicle.jpg4. Recommended Fastest Prototype Strategy
Section titled “4. Recommended Fastest Prototype Strategy”The fastest route is not to build the full system from scratch.
Recommended order:
Stage A — Prove the camera
Section titled “Stage A — Prove the camera”ANPR camera |record test footage |manually inspect platesGoal:
Moving night-time plates must first be readable by a human.
Stage B — Use existing open-source ANPR software
Section titled “Stage B — Use existing open-source ANPR software”Recommended first software:
Frigate NVR with built-in License Plate Recognition
This lets the project test:
RTSP camera |Frigate |vehicle detection |licence plate detection |OCR |history / searchwithout first writing the complete Python pipeline.
Stage C — Build custom Python code
Section titled “Stage C — Build custom Python code”Once the camera has been proven and real NZ plate samples have been collected:
OpenCV+ YOLO+ PaddleOCR+ custom NZ validation+ SQLite/PostgreSQLcan replace or complement Frigate.
Stage D — Build the remote solar node
Section titled “Stage D — Build the remote solar node”Only after Stage A/B works reliably.
This prevents spending money on:
- batteries;
- solar panels;
- 4G routers;
- cabinets;
- poles;
before knowing whether the camera can actually read plates.
5. System Architecture
Section titled “5. System Architecture”5.1 Initial local prototype
Section titled “5.1 Initial local prototype” ROAD================================================
Capture Zone | v [ Vehicle ] | v [ ANPR Camera ] | RTSP / LAN | v [ Test PC ] | +-------+-------+ | | v v [ Frigate ] [ Raw Video ] | v [ LPR ] | v [ Events / Search ]5.2 Final remote architecture
Section titled “5.2 Final remote architecture” SOLAR PANEL | v MPPT CONTROLLER | v LiFePO4 BATTERY | +-----------+-----------+ | | v v ANPR CAMERA 4G ROUTER | | [optional overview camera] | | | +------- Ethernet ------+ | v 4G / LTE | v Internet | v HOME / CLOUD SERVER | +-------------+-------------+ | | v v DATABASE DASHBOARD6. Prototype Phases
Section titled “6. Prototype Phases”Phase 1 — Camera proof
Section titled “Phase 1 — Camera proof”Use one ANPR camera near a location where power and Ethernet are temporarily available.
Do not use solar yet.
Test:
- daylight;
- night;
- headlights;
- 30 km/h;
- 50 km/h;
- rain;
- approaching vehicles;
- departing vehicles.
Success:
Most moving plates are manually readable.
Phase 2 — Existing software proof
Section titled “Phase 2 — Existing software proof”Install Frigate.
Connect the RTSP camera.
Enable LPR.
Success:
Camera -> Frigate -> recognised plate -> searchable historyPhase 3 — Custom Python experiment
Section titled “Phase 3 — Custom Python experiment”Use the same saved footage.
Build:
Video |YOLO vehicle detection |plate detector |PaddleOCR |NZ validation |CSV / SQLiteCompare custom recognition against Frigate.
Phase 4 — Remote solar prototype
Section titled “Phase 4 — Remote solar prototype”Move the camera to a location without mains power.
Add:
- battery;
- solar panel;
- MPPT controller;
- 4G router;
- weatherproof enclosure.
Phase 5 — Multi-camera network
Section titled “Phase 5 — Multi-camera network”Add Node 02, Node 03, etc.
Only do this after Node 01 has demonstrated acceptable reliability.
7. Camera Recommendation
Section titled “7. Camera Recommendation”Recommended prototype camera
Section titled “Recommended prototype camera”A strong candidate is:
Hikvision iDS-2CD7A26G0/P-IZHS — 8–32 mm
Why:
- purpose-built ANPR functionality;
- 2 MP progressive scan sensor;
- DarkFighter low-light capability;
- motorised 8–32 mm lens;
- optical zoom;
- built-in 850 nm IR;
- manual / semi-auto focus;
- shutter range to very fast exposures;
- RTSP/network streaming;
- weather-resistant body;
- onboard ANPR capability;
- suitable for narrow road capture zones.
For ANPR, a good 2 MP dedicated LPR camera can outperform a cheap 8 MP CCTV camera.
Example:
8 MP ordinary CCTV-------------------Street: excellentVehicle: excellentPlate: white / blurred
2 MP ANPR camera----------------Street: darkVehicle: darkPlate: ABC123The second image is more useful for this project.
8. Night-Time Number Plate Capture
Section titled “8. Night-Time Number Plate Capture”Night capture should use a dedicated plate capture zone.
Example:
CAPTURE ZONE | v
------------------------------------------------- [ ABC123 ] ----->------------------------------------------------- / / / [ ANPR Camera ]Recommended initial settings:
Resolution: 1920 x 1080Frame rate: 25–30 fpsShutter: 1/500 secondIR: ONIR wavelength: 850 nmGain: limitedWDR: test OFF initiallyFocus: fixed after setupCodec: H.264/H.265If motion blur remains:
1/500 |1/750 |1/1000Increase illumination rather than simply allowing longer exposure.
Camera position
Section titled “Camera position”Aim for:
Horizontal angle: preferably < 30 degreesVertical angle: preferably < 30 degreesUse optical zoom to make the plate occupy meaningful pixels.
Do not use one wide camera to perform both general surveillance and long-distance plate OCR.
9. Existing Software Options
Section titled “9. Existing Software Options”There are three sensible software strategies.
| Option | Cost | Setup Speed | Customisation | Recommended Use |
|---|---|---|---|---|
| Frigate | Free/open source | Fast | High | Best first prototype |
| OpenALPR open-source | Free | Medium/Hard | High | Experiments / reference |
| Custom YOLO + PaddleOCR | Free | Slowest | Maximum | Phase 2/3 development |
| Camera onboard ANPR | Included with ANPR camera | Very fast | Medium | Great hardware test |
| Commercial ANPR products | Paid | Fast | Varies | Later comparison |
The fastest prototype should combine:
Dedicated ANPR camera +FrigateFor the initial prototype, enable vehicle and person detection where useful. Treat face detection/cropping as optional, and keep automatic facial identity matching outside the baseline scope.
while also testing the camera’s own ANPR feature if available.
10. Recommended Open-Source Route — Frigate
Section titled “10. Recommended Open-Source Route — Frigate”Frigate is currently the best starting point for this prototype because it already provides:
- IP camera / RTSP support;
- video recording;
- object detection;
- event history;
- snapshots;
- retention rules;
- MQTT integration;
- built-in licence plate recognition;
- known plate matching;
- dedicated LPR camera mode.
Current Frigate LPR uses:
YOLOv9 licence plate detector +PaddleOCRand can operate locally after the models have been downloaded.
It also supports dedicated LPR cameras with a narrow field of view.
Why this is useful
Section titled “Why this is useful”Instead of initially building:
RTSP readervehicle trackingplate detectionOCRhistoryimage storagesearch UIyour first prototype can let Frigate provide most of those functions.
This means the first software experiment becomes:
1. Install Docker2. Install Frigate3. Add camera RTSP URL4. Enable LPR5. Define dedicated LPR camera6. Pass vehicles7. Check recognised platesThat could save a large amount of initial programming effort.
11. Alternative Open-Source / DIY Software
Section titled “11. Alternative Open-Source / DIY Software”11.1 OpenALPR
Section titled “11.1 OpenALPR”OpenALPR has an open-source library that can:
- analyse images;
- analyse video;
- return plate text;
- be called from C++, Python, C#, Java, Node.js and Go.
However, the open-source project is older and uses an AGPLv3 licence.
It is useful as:
- a reference implementation;
- a quick experiment;
- a comparison against a custom pipeline.
It would not be my first choice for this prototype in 2026.
11.2 YOLO + PaddleOCR
Section titled “11.2 YOLO + PaddleOCR”This is the recommended long-term custom route.
Pipeline:
RTSP / MP4 |OpenCV |YOLO vehicle/person detector |YOLO plate detector |plate crop |PaddleOCR |NZ plate validation |deduplication |databaseBenefits:
- complete control;
- can tune specifically for New Zealand plates;
- can use your own training data;
- can calculate familiarity/anomaly scores;
- integrates easily with your own C# backend.
Disadvantage:
- takes longer to build;
- more tuning;
- requires your own tracking and event logic.
12. Commercial / Ready-Made ANPR Software
Section titled “12. Commercial / Ready-Made ANPR Software”Commercial solutions are worth testing after the open-source proof of concept.
Possible categories:
Camera-native ANPR
Section titled “Camera-native ANPR”Some dedicated Hikvision and other professional ANPR cameras can perform recognition in the camera itself.
Advantages:
No GPU server required at remote siteLow bandwidthPlate metadata produced at cameraThis is especially useful for solar installations.
Commercial ANPR engines
Section titled “Commercial ANPR engines”Commercial engines may provide:
- better recognition accuracy;
- plate-country models;
- REST APIs;
- RTSP stream processing;
- support;
- easier deployment.
For this project, the correct strategy is:
Do not buy expensive software before testing Frigate and the camera’s built-in ANPR.
A commercial engine can later be run against the same recorded test set and objectively compared.
13. Custom Python ANPR Pipeline
Section titled “13. Custom Python ANPR Pipeline”Once the camera is proven, create a separate Python prototype.
Suggested components:
Python 3.11+OpenCVUltralytics YOLO (vehicle/person/plate detection)PaddleOCRSQLiteFastAPI (optional)Processing:
RTSP stream |frame sampling |vehicle detection |tracking |plate detection |best-frame selection |plate enhancement |OCR |normalisation |confidence scoring |deduplication |databaseNZ plate normalisation
Section titled “NZ plate normalisation”Common OCR confusion:
0 / O1 / I5 / S8 / BStore:
RawOCRNormalizedPlateConfidenceNever silently overwrite the raw OCR result.
14. Person Detection, Face Capture and Optional Face Recognition
Section titled “14. Person Detection, Face Capture and Optional Face Recognition”The roadside system can also detect people. This should be designed in levels so that the first prototype remains useful without immediately creating a biometric identification database.
14.1 Recommended capability levels
Section titled “14.1 Recommended capability levels”Level 1 — Person detection
Section titled “Level 1 — Person detection”Recommended for the first prototype.
Person detected: YESTimestamp: 01:14:28Camera: NODE-01Snapshot: person_011428.jpgThis allows an incident review to show that a person was present near a vehicle or capture zone.
Level 2 — Face detection / face crop
Section titled “Level 2 — Face detection / face crop”Optional after person detection is working.
Person detected |Face visible? |save best face cropThe system may save the best available face image associated with an incident or person event.
The purpose at this level is evidence preservation, not automatic identity matching.
Level 3 — Face recognition / identity matching
Section titled “Level 3 — Face recognition / identity matching”Not recommended as a Phase 1 requirement.
This level would compare a detected face with a stored biometric reference database and attempt to identify or match a person.
Example:
Face crop |face embedding / biometric template |reference database |possible matchThis creates substantially higher privacy, security, governance and accuracy requirements.
In New Zealand, automated facial recognition used to identify or categorise people can fall within the Biometric Processing Privacy Code 2025. If this feature is ever considered, complete a privacy impact assessment and confirm necessity, proportionality, safeguards, transparency, retention and access controls before implementation.
For this project, the recommended rule is:
Detect and preserve useful imagery first; do not attempt automatic identity recognition in the initial prototype.
14.2 Why a separate overview camera may be better
Section titled “14.2 Why a separate overview camera may be better”An ANPR camera is normally tuned for:
- fast shutter;
- narrow field of view;
- infrared illumination;
- reflective number plates;
- plate readability.
That can produce an excellent plate image while producing a poor person image.
Recommended future roadside node:
ROAD=====================================================
vehicle / person | +---------+---------+ | | v v ANPR / LPR camera Overview camera narrow field wider field fast shutter person/context plate exposure low-light image | | +---------+---------+ | v Frigate | +-------------+-------------+ | | | v v v Plate Vehicle PersonFor Prototype 1, one ANPR camera is enough.
For Prototype 2 or later, add an overview camera only if person/context imagery is genuinely useful.
14.3 Software approach for people
Section titled “14.3 Software approach for people”Frigate can already detect person objects when an appropriate object-detection model is configured.
A custom Python pipeline can use:
YOLO ├── car ├── truck ├── motorcycle └── personPossible event record:
EventIdTimestampCameraIdVehiclePresentPlatePersonPresentPersonSnapshotPathFaceCropPathDo not infer guilt from the presence of a person or vehicle. The output should be treated as investigative/search information only.
15. Remote Solar Camera Node
Section titled “15. Remote Solar Camera Node”A remote node should be a standalone roadside unit intended for a fixed location beside the road, preferably on private land or another specifically authorised site.
It should not depend on a nearby house for mains power or fixed broadband.
A remote node should be a standalone unit.
Solar Panel | v MPPT Controller | v LiFePO4 Battery | DC Fuse / Breaker | +-----------+-----------+ | | v v ANPR Camera LTE Router | | +------ Ethernet -------+Optional future addition:
Edge ComputerRaspberry Pi / Jetson / Mini PCBut do not add an edge computer initially unless required.
It consumes additional energy and adds maintenance complexity.
16. Power System Design
Section titled “16. Power System Design”Keep the node DC-only.
Avoid:
12V battery |230V inverter |AC adapter |12V devicebecause every conversion wastes energy.
Prefer:
12.8V LiFePO4 | +---- fused DC -> camera | +---- fused DC -> LTE routerA regulated DC/DC converter can be added where required.
17. Battery Sizing
Section titled “17. Battery Sizing”Recommended chemistry:
LiFePO4
Reasons:
- deep-cycle capable;
- long cycle life;
- stable voltage;
- lighter than lead-acid;
- high usable capacity;
- suitable for solar applications.
Avoid using an ordinary car starting battery.
Example power budget
Section titled “Example power budget”Conservative example:
ANPR camera 15 W4G router 3 Wconversion loss 2 W----------------------Total 20 WDaily energy:
20 W x 24 h= 480 Wh/day12.8 V 100 Ah battery
Section titled “12.8 V 100 Ah battery”Nominal:
12.8 V x 100 Ah= 1,280 WhAssuming ~80% usable:
~1,024 WhAt 20 W:
1,024 / 20= ~51 hoursSo a 100 Ah battery provides roughly two days of autonomy in this simplified example.
Recommended:
Prototype minimum: 100 AhBetter: 150 AhHigh resilience: 200 AhActual sizing must be recalculated after measuring real camera and router consumption.
Cold-weather note
Section titled “Cold-weather note”Use a battery with:
- low-temperature charge protection;
- suitable BMS;
- ideally a protected enclosure.
Do not allow a LiFePO4 battery to be charged outside its manufacturer’s safe temperature range.
18. Solar Panel Sizing
Section titled “18. Solar Panel Sizing”If consumption is approximately:
480 Wh/daya nominal 100 W panel is too marginal for a reliable 24/7 node.
Example theoretical output:
100 W x 4 peak-sun-hours= 400 Whand that is before real-world losses.
Recommended prototype sizing:
Minimum: ~200 WPreferred: ~250–300 WMore robust: ~300–400 WThe final design should account for:
- winter solar production;
- cloudy days;
- shading;
- panel orientation;
- charging efficiency;
- battery efficiency;
- dirt;
- system losses.
Use an MPPT controller rather than a basic PWM controller.
19. 4G / LTE Connectivity
Section titled “19. 4G / LTE Connectivity”Recommended class:
Industrial LTE router
Example:
Teltonika RUT241
Useful characteristics:
- 9–30 V DC input;
- Ethernet;
- LTE;
- external antenna support;
- low power consumption;
- suitable for unattended installations.
Published power figures are approximately around:
~1.3 W LTE idle~1.5 W with one LAN device~3.3 W under high traffic/loadActual field consumption must be measured.
Do not continuously stream video over 4G unless necessary
Section titled “Do not continuously stream video over 4G unless necessary”Preferred:
Camera records locally |ANPR event |plate metadata |small JPEG |4G uploadExample:
{ "camera": "NODE-01", "time": "2026-10-04T01:17:21", "plate": "ABC123", "confidence": 0.97, "direction": "eastbound"}Full video can remain on local SD storage and only be downloaded when an incident occurs.
20. Weatherproof Enclosure
Section titled “20. Weatherproof Enclosure”Use a locked IP65/IP66 outdoor enclosure.
Inside:
+--------------------------------+| Weatherproof Enclosure || || LiFePO4 battery || MPPT controller || fused distribution || DC/DC converter || LTE router || low-voltage protection || optional monitoring |+--------------------------------+Camera remains outside.
Solar panel remains outside.
LTE antenna may be mounted outside if required.
Add:
- cable glands;
- fused circuits;
- surge protection where appropriate;
- ventilation/thermal consideration;
- tamper-resistant fasteners;
- lock;
- labelled isolation point.
Electrical work should follow the requirements applicable to the actual installation.
21. Local vs Remote Processing
Section titled “21. Local vs Remote Processing”Option A — Camera onboard ANPR
Section titled “Option A — Camera onboard ANPR”Best for initial solar node.
Camera |onboard ANPR |plate + JPEG |4GAdvantages:
- lowest remote compute requirement;
- lower power;
- lower bandwidth;
- simpler remote hardware.
Option B — Home server processing
Section titled “Option B — Home server processing”Remote camera |4G RTSP |home server |Frigate / PythonAdvantages:
- centralised AI;
- easy software updates.
Disadvantages:
- higher data usage;
- needs reliable 4G;
- continuous RTSP consumes bandwidth.
Option C — Edge AI
Section titled “Option C — Edge AI”Camera |Jetson / Mini PC |ANPR |metadata onlyAdvantages:
- flexible;
- low network traffic.
Disadvantages:
- additional power;
- complexity;
- heat;
- maintenance.
Recommendation
Section titled “Recommendation”Prototype order:
1. camera onboard ANPR2. Frigate at home/local test network3. custom Python4. edge computer only if justified22. Networking and Security
Section titled “22. Networking and Security”Remote cameras must not simply expose RTSP or web administration directly to the Internet.
Preferred:
Remote node |LTE router |VPN tunnel |Home networkUse:
- WireGuard or equivalent VPN;
- unique passwords;
- HTTPS where available;
- firewall rules;
- no unnecessary inbound ports;
- firmware updates;
- separate camera VLAN at home;
- restricted admin access.
Example home network:
Camera VLAN192.168.30.0/24
ANPR server192.168.30.1023. Database Design
Section titled “23. Database Design”Camera
Section titled “Camera”Camera------IdNameDescriptionDirectionEnabledCreatedAtVehicleSighting
Section titled “VehicleSighting”VehicleSighting---------------IdCameraIdTimestampUtcPlateRawPlateNormalizedPlateConfidenceVehicleTypeVehicleColourDirectionPlateImagePathVehicleImagePathCreatedAtPlateStatistics
Section titled “PlateStatistics”PlateStatistics---------------PlateNormalizedFirstSeenLastSeenSeenCountNightSeenCountDaySeenCountTypicalHourTypicalDirectionUpdatedAtIncident
Section titled “Incident”Incident--------IdTitleStartTimeEndTimeDescriptionCreatedAt24. Common, Rare and New Vehicle Detection
Section titled “24. Common, Rare and New Vehicle Detection”Do not discard common vehicles.
Record sightings and classify familiarity.
Example:
ABC123Seen: 187Typical: 07:20–08:10Direction: EastboundStatus: COMMONAnother:
XYZ999Seen: 1First seen: 01:17Status: NEWA later anomaly score might use:
First ever appearance +40Seen fewer than 3 times +25Seen midnight–5am +20Unusual direction +10Outside normal time pattern +10Near incident time +20This is only a search prioritisation tool.
It must not be interpreted as evidence that a driver has committed an offence.
25. Incident Search Workflow
Section titled “25. Incident Search Workflow”If a vehicle theft occurs around 01:10:
Incident:00:45–01:30System query:
Show all platesbetween 00:45 and 01:30from Camera 1 / Camera 2sorted by rarityExample:
01:17 XYZ999 NEW 97%01:13 KLM456 RARE 95%01:08 ABC123 COMMON 98%Then inspect:
- vehicle image;
- plate image;
- direction;
- other cameras;
- full local footage.
26. Storage and Retention
Section titled “26. Storage and Retention”Suggested prototype policy:
Full video: 48–72 hoursVehicle snapshots: 14 daysPlate crops: 30 daysPlate metadata: 30 daysIncident evidence: preserved manuallyThese are engineering starting points, not legal requirements.
Implement automatic deletion.
If an incident occurs:
Create Incident |Select time window |Preserve linked media27. Privacy and Deployment Considerations
Section titled “27. Privacy and Deployment Considerations”Use conservative privacy practices.
Principles:
- capture only what is needed for security;
- avoid viewing into neighbours’ private areas;
- avoid recording audio unless genuinely necessary;
- do not publish plate histories;
- restrict system access;
- use secure passwords and VPN access;
- limit retention;
- preserve relevant footage for Police rather than public identification.
For the prototype, the safest practical approach is to mount the camera on private property.
Do not install poles, cabinets, solar equipment or cameras on council road reserve, berms, street furniture or other land without confirming permission from the relevant owner/authority.
28. Prototype Test Plan
Section titled “28. Prototype Test Plan”Build a repeatable dataset.
Target:
100–200 vehicle passesInclude:
| Condition | Test |
|---|---|
| Day | slow vehicle |
| Day | 30 km/h |
| Day | 50 km/h |
| Night | street lighting |
| Night | headlights toward camera |
| Night | rear plate |
| Night | wet road |
| Weather | rain |
| Traffic | two vehicles close together |
| Special | personalised plate |
| Special | dirty plate |
| Special | motorcycle if visible |
For every sample record:
Was plate manually readable?Was plate detected?Was OCR correct?Confidence?Vehicle direction correct?29. Success Criteria
Section titled “29. Success Criteria”Initial engineering targets:
Readable plate capture: >= 95%Correct OCR: >= 90%Readable plate capture: >= 90%Correct OCR: >= 85%The important distinction:
Capture failure != OCR failureTrack both separately.
30. Suggested Hardware List
Section titled “30. Suggested Hardware List”Prototype 1 — Local test
Section titled “Prototype 1 — Local test”| Item | Recommendation |
|---|---|
| Camera | Hikvision iDS-2CD7A26G0/P-IZHS 8–32 mm or equivalent dedicated ANPR camera |
| Network | Ethernet |
| Power | temporary mains / PoE for testing |
| Computer | existing Windows/Linux PC |
| Storage | existing PC storage |
| Software | Frigate first |
| Development | Python later |
Do not buy solar hardware yet.
Prototype 2 — Remote node
Section titled “Prototype 2 — Remote node”| Item | Recommended class |
|---|---|
| Camera | Dedicated 8–32 mm ANPR camera |
| SD card | 256–512 GB high-endurance |
| Router | Teltonika RUT241 or equivalent |
| Battery | 12.8 V 100–150 Ah LiFePO4 |
| Solar | ~250–300 W starting design |
| Controller | MPPT 20–30 A class |
| Enclosure | lockable IP65/IP66 |
| Protection | DC fuses / breakers |
| Network | LTE SIM + VPN |
| Overview/person camera | Optional later; not required for first ANPR test |
| Edge computer | None initially |
Final electrical sizing must be checked against actual measured loads and equipment specifications.
31. Suggested Software Stack
Section titled “31. Suggested Software Stack”Fast prototype
Section titled “Fast prototype”DockerFrigateSQLite / Frigate internal storageCustom development
Section titled “Custom development”PythonOpenCVUltralytics YOLO (vehicle/person/plate detection)PaddleOCRSQLiteFastAPIFuture production-style architecture
Section titled “Future production-style architecture”Python ANPR worker |REST / message queue |ASP.NET Core API |PostgreSQL |Vue / Angular frontendDo not start with the production architecture.
32. Recommended Repository Structure
Section titled “32. Recommended Repository Structure”anpr-prototype/│├── README.md├── docker-compose.yml│├── frigate/│ └── config.yml│├── python/│ ├── requirements.txt│ └── app/│ ├── capture/│ │ └── rtsp_capture.py│ ├── detection/│ │ ├── vehicle_detector.py│ │ └── plate_detector.py│ ├── ocr/│ │ └── plate_ocr.py│ ├── processing/│ │ ├── image_enhancement.py│ │ ├── plate_normalizer.py│ │ └── deduplicator.py│ └── main.py│├── data/│ ├── test-videos/│ ├── snapshots/│ ├── plates/│ └── ground-truth.csv│└── docs/33. Step-by-Step First Prototype
Section titled “33. Step-by-Step First Prototype”This is the recommended quickest route.
Step 1 — Buy/borrow one proper ANPR camera
Section titled “Step 1 — Buy/borrow one proper ANPR camera”Do not buy the entire remote system.
Use one dedicated ANPR-capable camera.
Step 2 — Temporary installation
Section titled “Step 2 — Temporary installation”Mount on private property.
Temporarily provide:
PoE / mainsEthernetAim at one narrow road zone.
Step 3 — Tune image capture
Section titled “Step 3 — Tune image capture”Before installing any OCR:
Test at night.
Tune:
zoomfocusshuttergainIRcamera anglecapture distanceStop here until plates are readable.
Step 4 — Save MP4 test footage
Section titled “Step 4 — Save MP4 test footage”Create your first benchmark dataset.
Example:
day_30kmh.mp4night_30kmh.mp4night_headlights.mp4rain_night.mp4Step 5 — Install Frigate
Section titled “Step 5 — Install Frigate”Install using Docker.
Add the camera’s RTSP stream.
First verify:
Live view worksRecording worksCar detection worksStep 6 — Enable LPR
Section titled “Step 6 — Enable LPR”Enable Frigate’s licence plate recognition.
For a dedicated plate camera, configure dedicated LPR mode.
Use saved debug plate images while tuning.
Step 7 — Compare recognition manually
Section titled “Step 7 — Compare recognition manually”Create:
timestamp,actual_plate,frigate_plate,correct21:03:11,ABC123,ABC123,true21:05:42,XYZ987,XYZ987,true21:07:01,DEF456,DEFG56,falseThis creates objective accuracy numbers.
Step 8 — Enable person detection
Section titled “Step 8 — Enable person detection”Once plate recognition is working, enable person detection in Frigate if the camera view includes an area where people may be relevant to an incident.
Verify:
person detectedsnapshot savedtimestamp correctDo not add face recognition at this stage.
If useful imagery of people cannot be obtained from the ANPR camera because the field of view is too narrow or exposure is optimised for reflective plates, plan an optional overview camera for the later remote node.
Step 9 — Test camera-native ANPR
Section titled “Step 9 — Test camera-native ANPR”If the chosen camera supports onboard ANPR:
compare:
Frigate resultvsCamera resulton the same vehicles.
Step 10 — Build a minimal Python test
Section titled “Step 10 — Build a minimal Python test”Do not build the whole backend.
First Python goal:
python recognize.py night_test.mp4Output:
timestamp,plate,confidenceUse:
YOLOPaddleOCRStep 11 — Decide software direction
Section titled “Step 11 — Decide software direction”Choose based on measured results.
If Frigate is good enough
Section titled “If Frigate is good enough”Keep Frigate and add your own API/analytics later.
If camera-native ANPR is better
Section titled “If camera-native ANPR is better”Let the camera recognise plates and send events centrally.
If neither is accurate enough
Section titled “If neither is accurate enough”Build/fine-tune the custom Python model.
This decision should be evidence-based.
34. Second Prototype — Remote Solar Node
Section titled “34. Second Prototype — Remote Solar Node”Only after the first prototype passes.
Build:
solar |MPPT |battery |+------------+| |camera LTE router| |+--- LAN ----+Then test:
- voltage at sunrise;
- voltage at sunset;
- daily Wh consumption;
- battery state of charge;
- solar harvest;
- cloudy-day performance.
Network
Section titled “Network”- LTE signal;
- connection reliability;
- VPN recovery;
- remote reboot.
Camera
Section titled “Camera”- same image quality as local test;
- local SD recording;
- ANPR event upload.
Run for at least several real weather cycles before adding a second node.
35. Future Multi-Camera Architecture
Section titled “35. Future Multi-Camera Architecture”Future:
NODE-01 --------\NODE-02 ---------\NODE-03 ----------> Central API -> PostgreSQL -> DashboardNODE-04 ---------/Each event:
{ "nodeId": "NODE-02", "timestamp": "...", "plate": "ABC123", "confidence": 0.97, "direction": "westbound"}This allows:
- same plate seen at different locations;
- time correlation;
- direction reconstruction;
- rare/new vehicle analysis.
Do not implement route profiling until there is a clear lawful and privacy-conscious need.
36. Features to Avoid Initially
Section titled “36. Features to Avoid Initially”Do not start with:
- facial recognition;
- automatic identity lookup;
- public plate database;
- neighbourhood-wide deployment;
- cloud microservices;
- Kubernetes;
- gRPC;
- custom mobile app;
- vehicle make/model AI;
- GPS tracking;
- complex anomaly AI;
- custom circuit boards.
They do not help answer the first engineering question.
37. Risks and Mitigations
Section titled “37. Risks and Mitigations”Risk — Plate unreadable at night
Section titled “Risk — Plate unreadable at night”Mitigation:
Better opticsFast shutter850 nm IRNarrow capture zoneSmaller angleRisk — Battery dies in winter
Section titled “Risk — Battery dies in winter”Mitigation:
Measure real powerOversize batteryOversize solarMPPTLow-voltage cutoffRemote battery telemetryRisk — LTE data cost
Section titled “Risk — LTE data cost”Mitigation:
Local recordingMetadata-only uploadsJPEG event uploadDownload full video only after incidentsRisk — Camera/node stolen
Section titled “Risk — Camera/node stolen”Mitigation:
- mount out of easy reach;
- locked metal enclosure;
- tamper-resistant fasteners;
- hidden identification;
- offline alert;
- optional tracker later.
Risk — OCR false positives
Section titled “Risk — OCR false positives”Mitigation:
- retain original plate image;
- store confidence;
- preserve raw OCR;
- require manual verification before treating any plate as relevant.
Risk — Privacy concerns
Section titled “Risk — Privacy concerns”Mitigation:
- narrow capture zone;
- short retention;
- restricted access;
- no public publication;
- use for security incidents;
- confirm permissions for any non-private-property installation.
38. Final Recommended Build Order
Section titled “38. Final Recommended Build Order”Recommended sequence:
1. Buy ONE proper ANPR camera |2. Temporary powered installation |3. Tune nighttime plate capture |4. Record benchmark videos |5. Install Frigate |6. Test built-in Frigate LPR |7. Test camera-native ANPR |8. Enable/test person detection |9. Measure recognition accuracy |10. Build minimal Python comparison |11. Choose recognition engine |12. Build roadside solar + battery node |13. Add LTE + VPN |14. Run Node-01 for several weeks |15. Add optional overview/person camera if justified |16. Add database / anomaly analytics |17. Consider face cropping only if useful |18. Consider biometric face matching only after privacy/legal review |19. Only then consider more roadside nodesThis is deliberately different from trying to build everything at once.
It gives the fastest answer to:
Does this idea actually work under real local night-time road conditions?
39. Definition of Done
Section titled “39. Definition of Done”The first complete prototype is successful when:
- a moving vehicle enters the capture zone;
- the plate is physically readable in the camera image;
- the software detects it;
- OCR produces the correct plate;
- duplicate frames are collapsed into one vehicle event;
- the event receives a timestamp;
- direction is recorded if available;
- a plate crop is saved;
- a vehicle image is saved;
- the sighting can be found by time;
- night performance is acceptable;
- person detection can save a timestamped person snapshot where the camera view supports it;
- automatic face identity recognition is not required for Prototype 1;
- data is automatically expired according to retention settings.
The remote prototype is successful when all the above works while:
- powered only by battery/solar;
- connected only by LTE/4G;
- surviving multiple days of normal weather;
- remotely accessible through a secure network connection;
- recovering automatically after network or power interruptions.
40. Reference Links
Section titled “40. Reference Links”These references should be rechecked before purchasing hardware because software and product specifications can change.
Frigate
Section titled “Frigate”- Documentation: https://docs.frigate.video/
- Licence Plate Recognition: https://docs.frigate.video/configuration/license_plate_recognition/
- GitHub: https://github.com/blakeblackshear/frigate
Frigate currently supports local LPR, dedicated LPR camera configuration, recognised plate fields, history filtering and MQTT events.
Hikvision ANPR camera
Section titled “Hikvision ANPR camera”- Hikvision iDS-2CD7A26G0/P-IZHS(Y) product information: https://hsrc.hikvision.com/en/products/IP-Products/Network-Cameras/DeepinView-Series/ids-2cd7a26g0-p-izhs-y-/
Useful specifications include the 8–32 mm lens option, 850 nm IR and dedicated licence-plate recognition support.
OpenALPR
Section titled “OpenALPR”The open-source library is licensed under AGPLv3. Review licence requirements carefully before incorporating it into any commercial product.
Teltonika RUT241
Section titled “Teltonika RUT241”- Product / documentation: https://teltonika-networks.com/products/routers/rut241
- Power consumption: https://wiki.teltonika-networks.com/view/RUT241_Power_Consumption
New Zealand privacy guidance
Section titled “New Zealand privacy guidance”-
CCTV guidance — Office of the Privacy Commissioner: https://www.privacy.org.nz/resources-and-learning/a-z-topics/privacy-and-cctv/
-
Biometric Processing Privacy Code 2025: https://www.privacy.org.nz/privacy-principles/codes-of-practice/biometric-processing-privacy-code/
The Biometric Processing Privacy Code is especially relevant if the system later performs automated facial identity matching or biometric categorisation.
Practical Recommendation
Section titled “Practical Recommendation”For the very first experiment, build only this before placing anything at a remote roadside location:
Hikvision ANPR camera | Ethernet | Existing PC | FrigateNo solar.
No 4G.
No custom dashboard.
No C# backend.
No Raspberry Pi.
First prove:
Night moving vehicle |clear plate image |correct plate recognitionOnce this works, the second experiment is a single authorised roadside node:
Roadside ANPR camera |optional overview/person camera |LiFePO4 + Solar |LTE router |VPN / remote accessOnly after the remote node is stable should the project expand into a multi-location neighbourhood system.
That approach provides the fastest, lowest-risk path from idea to a working ANPR prototype.