AI Agent Hub
Back to skills
Tencent Cloud Vehicle License OCR icon

Tencent Cloud Vehicle License OCR

Development Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @tencent-adm/tencentcloud-ocr-vehiclelicense according to https://skillhub.cn/install/skillhub.md.

About this skill

Problem

Vehicle license images often vary by front/back page, electronic version, tractor license, or copied/screenshot source. Manual extraction is error-prone, and review systems need signals for quality issues such as blur, glare, or incomplete borders.

How It Works

The skill wraps the Tencent Cloud OCR VehicleLicenseOCR API. A request can provide ImageBase64 or ImageUrl, and set CardSide to FRONT, BACK, or DOUBLE; tractor licenses can use TractorCardSide for front or back. It returns JSON fields such as plate number, vehicle type, owner, VIN, engine number, and registration date for the front; archive number, approved passenger count, total mass, curb weight, and external dimensions for the back. Electronic licenses also include status, inspection validity, and generation time.

Boundaries

  • The image should be PNG/JPG/BMP; Base64 should stay under 7M; URL download should finish within 3 seconds.
  • CardSide must match the actual card side, or the API may return FailedOperation.LicenseCardSideError.
  • Alert codes can coexist, covering copies, screenshots, glare, blur, and incomplete borders.
  • The account must have the service enabled and valid quota; non-license images, empty images, multiple cards, arrears, or exhausted packages produce errors.

Use Cases

  • Extract plate number, vehicle type, owner, and VIN from front license pages during owner onboarding.
  • Read archive number, approved passengers, total mass, and load capacity from back license pages in used-car review.
  • Detect copied, screenshot, glare, or blur issues in license images during document quality checks.
  • Recognize tractor license back pages and return structured fields for agricultural vehicle systems.

Best For

  • Backend engineers for vehicle services who need structured license fields from OCR results.
  • Review engineers who need to flag copies, screenshots, blur, glare, or incomplete borders.
  • Engineers building tractor or used-car systems who need front, back, or double-side recognition.
  • Python developers integrating Tencent Cloud APIs and handling image inputs, alerts, and error codes.