ANPR + MMR Demo

Web Demo

Usage

REST API

Register new developer

The registration is needed for using REST API. Type your email address and you will obtain a password.

REST API Documentation

REST Interafce

API uses REST-like interface. This means that all calls to the API are made over the Internet, using HTTP POST requests to the API server (https://cloud.eyedea.cz/api/v2/mmrdetect). Any programming language that can be used to communicate over HTTP can be used for client-side implementations.

Responses are returned in JSON format.

usage notes:

All coordinates and sizes are provided in pixels.

The maximum image size you can treat with is 20MB.

Download python api client

Get server system info
URL/v2/serverSystemInfo HTTP MethodPOST
Consumes Media Type- Produces Media Typeapplication/json
URL Parameters-
CURL Command Example curl -X POST -F "email=your_email" -F "password=your_password"
https://cloud.eyedea.cz/api/v2/serverSystemInfo
Data Parameters email - the email provided by user during registration
password - the user's password obtained after registration.
Example Response

Success

{
"timestamp": 1611653049425,
"systemLoadAverage": 8.9,
"systemUsedMemory": 6.25,
"systemUsedMemoryBytes": 2147483648,
"processingServersInfosList": [
{,
"batchSize": 1,
"computationMode": "CPU",
"engineName": "LPM",
"engineType": "Detector",
"engineVersion": "v7.1-2020-04-16-Ubuntu-16.04-hasp",
"moduleId": 2,
"moduleVersion": "002-frontal.lp-eu-v7.5",
"numComputingThreads": 1,
"numWaitingTasks": 0,
"status": "RUNNING"
},
{
"batchSize": 1,
"computationMode": "GPU",
"engineName": "LPM",
"engineType": "ANPR",
"engineVersion": "v7.1-2020-04-16-Ubuntu-16.04-hasp",
"moduleId": 2,
"moduleVersion": "002-frontal.lp-eu-v7.5",
"numComputingThreads": 1,
"numWaitingTasks": 0,
"status": "RUNNING"
},
{
"batchSize": 1,
"computationMode": "GPU",
"engineName": "Eyedentify",
"engineType": "MMR",
"engineVersion": "2.9.0-Ubuntu-16.04-x86_64-HASP",
"moduleName": "MMR_VCMM_PREC_2020Q4.dat",
"moduleVersion": "20201111",
"numComputingThreads": 1,
"numWaitingTasks": 0,
"status": "RUNNING"
},
{
"batchSize": 1,
"computationMode": "GPU",
"engineName": "Eyedentify",
"engineType": "MMR Color",
"engineVersion": "2.9.0-Ubuntu-16.04-x86_64-HASP",
"moduleName": "MMR_COLOR_PREC_2020Q4.dat",
"moduleVersion": "20201129",
"numComputingThreads": 1,
"numWaitingTasks": 0,
"status": "RUNNING"
}
],
"gpuInfoList": [
{
"id": 0,
"name": "GeForce GTX TITAN X",
"totalMemoryMB": 12288,
"freeMemoryMB": 11776,
"gpuUtilizationPerc": 0
}
]
}
Detect license plates in image and get results for ANPR and MMR
URL/api/v2/mmrdetect HTTP MethodPOST
Consumes Media Typemultipart/form-data Produces Media Typeapplication/json
URL Parameters-
Data Parameters email - the email provided by user during registration
password - the user's password obtained after registration.
file - form data parameter containing input image.
lpDetection - (optional) form data parameter containing JSON with license plate positions (for MMR computation only).

Input parameter lpDetection contains the JSON string defining the array containing the positions of the license plates in the input image. Structure Center, angle, scale is used to define the position of the license plate in the image.

Note: If the parameter lpDetection is used, automatic detection and ANPR will not run during the input image processing.

WARNING: The scale defines the real size of the license plate in the image and does not take into account the margin added by the detector.

Center, angle, scale 
[
{
"center": {
"x": 176.5,
"y": 362.0
},
"angle": -5.1,
"scalePPM": 214.3
},
...
]

Structure Center, angle, scale defines the position of the vehicle. The structure contains the 2D point center defining the center of the license plate, property angle defining the image correction angle (CW correction of the source image in degrees) and the property scalePPM defining the real world scale (in pixels per meter) of the input image in the center position.

The center point must be inside the input image.
CURL Command Example Use automatic detection:
     curl -X POST
     -H "Content-Type: multipart/form-data"
     -F "file=@image.jpg"
     -F "email=your_email"
     -F "password=your_password"
     https://cloud.eyedea.cz/api/v2/mmrdetect
Detections in JSON string:
     curl -X POST
     -H "Content-Type: multipart/form-data"
     -F "file=@image.jpg"
     -F "email=your_email"
     -F "password=your_password"
     -F "lpDetection=[{\"center\":{\"x\":176.5,\"y\":362.0},\"angle\":-5.1,\"scalePPM\":214.3}]"
     https://cloud.eyedea.cz/api/v2/mmrdetect
Notes:
  • Content type of the request is multipart/form-data, where the inputs are stored in the form data fields.
  • Input image is referenced with filepath (starting with @) in the form field file.
  • Optional parameter lpDetection can contain the whole JSON as a string (special characters must be escaped).
Response Codes

Success

Example Content: 
{
"status":success,
"width":2400,
"height":1600,
"libraryVersion":{
"LPM ANPR Engine":{
"moduleVersion":
 "002-frontal.lp-eu-v7.5",
"version":
 "v7.1-2020-04-16-Ubuntu-16.04-hasp"
},
"LPM Detector Engine":{
"moduleVersion":
 "002-frontal.lp-eu-v7.5",
"version":
 "v7.1-2020-04-16-Ubuntu-16.04-hasp"
},
"MMR Color Engine":{
"moduleVersion":"20201129",
"version":
 "2.9.0-Ubuntu-16.04-x86_64-HASP"
},
"MMR Engine":{
"moduleVersion":"20201111",
"version":
 "2.9.0-Ubuntu-16.04-x86_64-HASP"
}
},
"tags":[
{
"lpDetection":{
"corners": [
{
"x": 1234.56,
"y": 234.56
},
{
"x": 1289.10,
"y": 234.56
},
{
"x": 1289.10,
"y": 244.67
},
{
"x": 1234.56,
"y": 244.67
}
],
"width": 54.54,
"height": 10.11,
"angles": [ 5.0, 0.0, 0.0 ],
"score": 0.7891011,
"position": {
"center": {
"x": 1261.83,
"y": 239.615
},
"angle": -5.0,
"scalePPM": 93.65
}
},
"anprResult":{
"country": "CZ",
"countryScore": 0.987654321,
"ocrText": "1A23456",
"ocrTextScore": 0.953135953
},
"mmrResult":{
"category": "CAR",
"categoryID": 2,
"categoryScore": 0.9999992,
"color": "GRAY",
"colorID": 7,
"colorScore": 0.9012345,
"make": "VW",
"makeID": 43,
"makeScore": 0.9999914,
"model": "Golf",
"modelID": 3544,
"modelScore": 0.9234567,
"view": "frontal",
"viewID": 1,
"viewScore": 0.9999999
}
},
...
]
}

Error

  • Code: 20 IMG_DECODE_ERROR
  • Code: 29 UPLOAD_ERROR
  • Code: 31 FILE_NOT_FOUND
  • Code: 33 FILE_TOO_LARGE
  • Code: 34 MALFORMED_URL
  • Code: 104 INTERNAL_ERROR
  • Code: 204 INVALID_LOGIN
  • Code: 402 MISSING_ARGUMENTS
  • Code: 205 DAILY_API_USAGE_LIMIT_EXCEEDED