🚗 AI License Plate Recognizer (OCR)

Read vehicle license plates from image URLs using AI vision. Send one or more public image links (JPG, PNG, WebP and more) and get a clean JSON record per image: the plate number, a confidence score, the vehicle make/model/color, whether it's day or night, image metadata, and a reason when a plate can't be read (dirty plate, glare, rain, blur...).

🎯 Built for plates. Instead of a generic OCR that dumps raw text, this Actor is prompted specifically to find and normalize the license plate and describe the vehicle and scene around it.

---

What you get

	Field	Description
🔑	licensePlate	The recognized plate, uppercased and normalized (main field)
🎯	plateConfidence	Confidence of the read (0-1)
🌍	plateRegion	Plate country/standard when identifiable (e.g. "Brazil (Mercosul)")
🌗	timeOfDay	day, night or unknown
🌦️	weather	Scene weather: clear, rain, fog, snow, cloudy, unknown
🔍	imageQuality	Perceived quality: good, fair, poor
📐	viewAngle	Camera view: front, rear, side, unknown
🚙	vehicleMake / Model / Color / Type	Vehicle attributes (type: car, motorcycle, truck...)
🖼️	imageWidth / imageHeight / imageFormat / imageSizeBytes	Image pixel size, format and byte size (flat, one column each)
⚠️	failureReason	Why the plate couldn't be read, when applicable

---

How to use

1. Provide image URLs

{
  "imageUrls": [
    "https://upload.wikimedia.org/wikipedia/commons/6/66/Car_vehicle_israel_license_plate_%284893367447%29.jpg",
    "https://upload.wikimedia.org/wikipedia/commons/1/19/Michigan_Dealer_License_Plate.jpg"
  ]
}

Supports JPG/JPEG, PNG, WebP and other common image formats. Up to 50 images per run (send larger volumes via sequential calls).

2. Run it - each image is downloaded and analyzed in parallel.

3. Read the JSON - one record per image in the dataset (export as JSON, CSV or Excel), or get it synchronously via the Standby API.

---

Pricing

Charged per image where a plate is read (event plate-recognized).

• Images that fail to download: not charged
• Images where no plate is readable: not charged

---

Example output

[
  {
    "imageUrl": "https://.../car.jpg",
    "success": true,
    "licensePlate": "ABC1D23",
    "plateConfidence": 0.97,
    "plateRegion": "Brazil (Mercosul)",
    "timeOfDay": "day",
    "weather": "clear",
    "imageQuality": "good",
    "viewAngle": "front",
    "vehicleMake": "Volkswagen",
    "vehicleModel": "Gol",
    "vehicleColor": "silver",
    "vehicleType": "car",
    "imageWidth": 1280,
    "imageHeight": 720,
    "imageFormat": "jpg",
    "imageSizeBytes": 184320,
    "failureReason": null,
    "processedAt": "2026-01-01T12:00:00.000Z",
    "error": null
  },
  {
    "imageUrl": "https://.../blurry.jpg",
    "success": false,
    "licensePlate": null,
    "plateConfidence": null,
    "plateRegion": null,
    "timeOfDay": "night",
    "weather": "rain",
    "imageQuality": "poor",
    "viewAngle": "rear",
    "vehicleMake": null,
    "vehicleModel": null,
    "vehicleColor": "black",
    "vehicleType": "car",
    "imageWidth": 800,
    "imageHeight": 600,
    "imageFormat": "jpg",
    "imageSizeBytes": 95000,
    "failureReason": "glare",
    "processedAt": "2026-01-01T12:00:01.000Z",
    "error": null
  }
]

---

FAQ

Which images work best? Clear photos where the plate is visible. Heavy glare, motion blur, dirt, rain or very low resolution may prevent a read - the reason is reported in failureReason.

What plate formats are supported? The model reads plates from most regions and reports the standard in plateRegion when it can identify it.

Can I call it in real time? Yes. The Standby endpoint POST /recognize responds synchronously. See below.

Do failed images cost anything? No. You are only charged for images where a plate is actually read.

---

🔌 API integration

Batch run:

curl -X POST "https://api.apify.com/v2/acts/acme-ai~plate-recognizer-ai/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"imageUrls":["https://upload.wikimedia.org/wikipedia/commons/6/66/Car_vehicle_israel_license_plate_%284893367447%29.jpg","https://upload.wikimedia.org/wikipedia/commons/1/19/Michigan_Dealer_License_Plate.jpg"]}'

Standby (POST /recognize):

curl -X POST "https://acme-ai--plate-recognizer-ai.apify.actor/recognize" \
  -H "Authorization: Bearer YOUR_APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  --compressed \
  -d '{"imageUrls":["https://upload.wikimedia.org/wikipedia/commons/6/66/Car_vehicle_israel_license_plate_%284893367447%29.jpg","https://upload.wikimedia.org/wikipedia/commons/1/19/Michigan_Dealer_License_Plate.jpg"]}'

The token goes in the Authorization: Bearer header, never in the URL.

---

Notes

This Actor analyzes images you provide. You are responsible for having the right to process the images and any personal data they may contain (such as plates or people).