License Plate Recognition API Guide

Number Plate Detection API

Automatically identify and read vehicle licence plates from an image, with a bounding box and confidence score for every plate found.

Select version

Download document

Postman Swagger

Overview

The Number Plate Detection API identifies and reads vehicle licence plates from an image. Built on machine learning models tuned for real-world capture conditions, it suits traffic management, security and automation workloads.

Details

Request URL

https://faceapi.mxface.ai/api/Vehicle/NumberPlateDetect

Method type

POST

Headers

Content-Type : application/json

Subscriptionkey : xxxxxxxxxxx

Request body

{
  "encoded_image": "string"
}

Request parameters

encoded_image YesString

Base64 encoded string of image.

Response body

[
  {
    "code": 0,
    "message": "string",
    "errorMessage": "string",
    "numberPlates": [
      {
        "boundingBox": {
          "x": "number",
          "y": "number",
          "height": "number",
          "width": "number"
        },
        "confidence": "float",
        "numberPlateText": "string",
        "numberPlateTextConfidence": "float"
      }
    ]
  }
]

Response parameters

code int

Status code.

message string

Status message.

errorMessage string

A human readable description of the error.

numberPlates array

List of detected number plates.

boundingBox object

Bounding box of the detected plate.

x int

x-edge of the plate.

y int

y-edge of the plate.

height int

Height of the plate from y.

width int

Width of the plate from x.

confidence float

Confidence of the detection.

numberPlateText string

Number plate text.

numberPlateTextConfidence float

Confidence of the detected plate text.

Errors

200

Success

400

Invalid request. Please pass valid json with all required parameters.
Not valid biometric data or invalid format.
The input is not a valid Base-64 string.

401

Access denied due to an invalid subscription key or the wrong API endpoint. Make sure to provide a valid key for an active subscription.

500

Something went wrong, please try again later.