PlateVision SDK

Real-time, offline license plate recognition for iOS, Android & Flutter.

🍎 iOS 16+ 🤖 Android 8+ 💙 Flutter 3+ ✈️ Offline 🌎 90+ Countries

PlateVision is a mobile SDK delivering accurate, real-time ALPR entirely on-device. No cloud round-trips, no per-scan cost, no connectivity required while scanning.

ℹ️

You need an API key. Create a free account — no credit card required.

Quickstart

Scan plates in your iOS app in under 5 minutes.

Get your credentials
Sign up and copy your API key and secret from the dashboard.
Build and add the framework
Run ./build.sh from the Framework folder, drag the resulting PlateVisionSDK.xcframework into Xcode, set Embed & Sign.
Add camera permission
Add NSCameraUsageDescription to Info.plist.
Activate and scan

swift

import PlateVisionSDK

PlateVisionSDK.shared.activate(apiKey: "pv_live_...", secretKey: "sk_...") { result in
    guard case .success(let l) = result, l.isValid else { return }
    PlateVisionSDK.shared.delegate = self
    PlateVisionSDK.shared.startScanning()
}

func plateVision(_ sdk: PlateVisionSDK, didDetect result: PVDetectionResult) {
    print(result.plateText, result.confidencePercent)
}

💡

Add the preview. Use PVCameraPreview(sdk: .shared) in SwiftUI or access PlateVisionSDK.shared.previewLayer for UIKit.

Authentication

Every SDK call is authenticated with an API key + secret key pair from your dashboard.

Credential	Format	Notes
`apiKey`	pv_live_...	Identifies your app. Safe to bundle in the binary.
`secretKey`	sk_...	Authenticates activation. Keep private, never log.

⚠️

Never commit keys to source control. Use environment variables or a gitignored config file.

iOS SDKiOS

Installation

Binary XCFramework for iOS 16+ (arm64 device + arm64/x86_64 Simulator).

bash — build the framework

cd PlateVisionSDK/Framework
chmod +x build.sh && ./build.sh
# Outputs: build/PlateVisionSDK.xcframework

Then drag PlateVisionSDK.xcframework into Xcode → set Embed & Sign → add NSCameraUsageDescription to Info.plist.

Swift Package Manager

Package.swift

.package(url: "https://github.com/platevision/platevision-ios-sdk", from: "1.0.0")

Activation

swift

PlateVisionSDK.shared.activate(apiKey: key, secretKey: secret) { result in
    switch result {
    case .success(let l) where l.isValid: // ready
    case .failure(let e): print(e)
    default: break
    }
}

Configuration

swift

var config = PVConfiguration.default
config.region              = .us
config.minimumConfidence   = 0.70
config.deduplicationWindow = 3.0
config.enableSound         = true
config.scanROI = CGRect(x:0.07, y:0.40, width:0.86, height:0.20)
PlateVisionSDK.shared.configure(config)

Property	Type	Default	Description
`region`	PVRegion	.auto	Country hint for validation.
`minimumConfidence`	Float	0.60	OCR threshold (0–1).
`deduplicationWindow`	TimeInterval	2.0 s	Suppress duplicate detections.
`enableSound`	Bool	true	System tick on each hit.
`scanROI`	CGRect	centre 86×20%	Normalised pixel-buffer crop rect.

Delegate

swift

public protocol PlateVisionDelegate: AnyObject {
    func plateVision(_ sdk: PlateVisionSDK, didDetect result: PVDetectionResult)
    func plateVision(_ sdk: PlateVisionSDK, didChangeState state: PVState)
    func plateVisionDidFailLicence(_ sdk: PlateVisionSDK)
}

PVDetectionResult	Type	Description
`plateText`	String	Uppercased stripped plate number
`confidence`	Float	OCR confidence 0–1
`confidencePercent`	Int	confidence × 100
`boundingBox`	CGRect	Normalised Vision rect within scan ROI
`timestamp`	Date	Time of detection
`region`	PVRegion	Active region at time of detection

Camera UI

swift — SwiftUI

ZStack {
    PVCameraPreview(sdk: .shared).ignoresSafeArea()
    // overlay: scan guide, result card, controls...
}

swift — UIKit

if let layer = PlateVisionSDK.shared.previewLayer {
    layer.frame = view.bounds
    view.layer.insertSublayer(layer, at: 0)
}

Full API Reference

Method / Property	Description
`activate(apiKey:secretKey:completion:)`	Validate credentials. Once per launch.
`configure(_ config:)`	Apply config any time.
`startScanning()`	Request camera permission + start.
`stopScanning()`	Stop and tear down session.
`pause()` / `resume()`	Suspend/restore without teardown.
`captureOnce()`	Detect one plate then auto-pause.
`delegate`	PlateVisionDelegate receiver.
`state`	.idle / .running / .paused / .permissionDenied / .error
`previewLayer`	AVCaptureVideoPreviewLayer (set after startScanning).
`licenceInfo`	PlateVisionLicence (set after activation).

Android SDKAndroid

build.gradle.kts

dependencies { implementation("io.platevision:sdk-android:1.0.0") }

kotlin

PlateVisionSDK.getInstance().activate(context=this, apiKey="pv_live_...", secretKey="sk_...") { result ->
    if(result.isSuccess) {
        PlateVisionSDK.getInstance().startScanning(lifecycleOwner=this, previewView=binding.pv,
            onDetection = { p -> Log.d("PV", p.plateText) })
    }
}

Flutter SDKFlutter

pubspec.yaml

dependencies:
  platevision_sdk: ^1.0.0

dart

final l = await PlateVisionSDK.activate(apiKey:'pv_live_...', secretKey:'sk_...');

PlateVisionCamera(
  onDetection: (r) => print(r.plateText),
  config: PVConfig(region: PVRegion.auto, minimumConfidence: 0.7),
)

Regions

Region	Enum	Length	Example
Auto (global)	`.auto`	5–10	Any valid plate
US / Canada	`.us`	5–8	`ABC1234`, `353-42T`
United Kingdom	`.uk`	5–7	`AB12CDE`
Europe	`.europe`	5–8	`AB1234CD`
Australia	`.australia`	5–7	`ABC123`
India	`.india`	8–10	`MH01AB1234`
Middle East	`.middleEast`	5–8	`ABC12345`

Plate validation pipeline

Clean — uppercase, strip spaces, dashes, dots
Length — per-region min/max bounds
Alphanumeric — reject symbols
Mix check — must have ≥1 letter AND ≥1 digit
Run check — reject 4+ identical consecutive chars
False-positive list — block common OCR noise words
Region regex — country-specific pattern (when region is not .auto)

Region of interest

scanROI is a normalised rect (0–1, origin top-left) used to physically crop the pixel buffer before OCR. Only pixels inside this rectangle are passed to Vision — text outside is genuinely invisible to the engine.

swift

// Default: 86% wide, 20% tall, vertically centred
config.scanROI = CGRect(x:0.07, y:0.40, width:0.86, height:0.20)

Offline mode

All OCR inference runs on-device. Licence validation only needs internet once — on first activation. Results are cached; the SDK is fully offline thereafter.

💡

Critical for parking garages, tunnels, and any environment where connectivity is unreliable.

Changelog

v1.0.0 — March 2026 Latest

Initial public release
iOS SDK: AVFoundation + Vision, fully offline
7-stage plate validation with per-region patterns
Pixel buffer ROI cropping
7 regions / 90+ country support
SwiftUI PVCameraPreview + UIKit preview layer
API key + secret activation with offline fallback
Configurable confidence, deduplication, sound, ROI
Single-shot captureOnce() mode

Support

Email: support@platevision.io
Get started: Create a free account
GitHub: github.com/platevision

Pro & Enterprise: same-day response. Starter: 2–3 business days.