Blur API Reference

Welcome to the Plate Recognizer Blur API! Use our API to blur license plates and faces in images.
Check detailed instructions on how to install Blur SDK.
Blur forwards images to Snapshot API then uses the detected plate bounding box to perform accurate blurring.

Authentication

No authentication is required to access Blur API endpoints

Blur Endpoint

POST http://localhost:8001

POST Parameters

Parameter	Required	Description
upload	Yes	The file to be uploaded. optional if upload_url param is present.
upload_url	No	The url of file to be uploaded. Given priority over upload param.
regions	No	Regions forwarded to Snapshot
skip_regexp	No	Override configuration SKIP_REGEXP. Can be specified multiple times.
faces	No	Override configuration FACES.
plates	No	Override configuration PLATES
logo	No	Logo file. If sent, Logo file path is ignored.
logo	No	Logo file path. Overrides configuration LOGO
config	No	Extends configuration CONFIG.
et	No	Exclude a percentage margin of image height at the top.
el	No	Exclude a percentage margin of image width at the left.
eb	No	Exclude a percentage margin of image height at the bottom.
er	No	Exclude a percentage margin of image width at the right.
xmin	No	Excluded region bounding box xmin.
ymin	No	Excluded region bounding box ymin.
xmax	No	Excluded region bounding box xmax.
ymax	No	Excluded region bounding box ymax.
split	No	Split image into horizontal chunks before blur. Use on panoramic images
overlap	No	Overlap percentage if splitting image. Default is 20.
camera_id	No	Override configuration CAMERA_ID.
snapshot_filename	No	Custom file name sent to Snapshot, Default is the uploaded file name.

Request Format

Shell

# Process an image from a URL
curl -o /tmp/blur-car.jpg -d 'upload_url=https://app.platerecognizer.com/static/demo.jpg' -d 'faces=10' -d 'plates=10' http://localhost:8001/

# Get back unskewed bounding box in JSON format and skip burring
curl -d 'upload_url=https://app.platerecognizer.com/static/demo.jpg' http://localhost:8001/

# Download a test picture
curl -o /tmp/car.jpg https://app.platerecognizer.com/static/demo.jpg
# Upload to API for blurring
curl -F 'upload=@/tmp/car.jpg' -o /tmp/blur-car.jpg -F 'faces=10' -F 'plates=10' http://localhost:8001/
# Upload image and logo
curl -F 'upload=@/tmp/car.jpg' -F 'logo=@/tmp/logo.jpg' -o /tmp/blur-car.jpg -F 'faces=10' -F 'plates=10' http://localhost:8001/

# Curl on Windows
curl -X POST -F upload=@car.jpg -o blur-car.png http://localhost:8001/

Python

import requests

files = {
    'upload': open('/tmp/car.jpg', 'rb'),
    'plates':'10',
    'faces':'10',
}

response = requests.post('http://localhost:8001/', files=files)

with open('/tmp/blur-car.jpg', 'wb') as f:
    f.write(response.content)

Response Formats

There are two possible response formats depending on the configuration and request params.

1. Polygon points around the plates and faces in JSON. This is the default format.
2. Blurred image if you include faces or plates params or set a blur intensities more than 0 in configuration.

Response Schema

{
  "faces": [
    {
      "box": {
        "xmax": 326,
        "xmin": 300,
        "ymax": 137,
        "ymin": 100
      },
      "score": 0.7088521122932434
    }
  ],
  "plates": [
    {
      "polygon": [
        [
          157.375,
          494.484375
        ],
        [
          272.546875,
          544.453125
        ],
        [
          268.890625,
          581.625
        ],
        [
          153.109375,
          531.65625
        ]
      ],
      "result": {}
    }
  ]
}

Attribute	Description
faces/box	Face detection bounding box
faces/score	Face detection score between 0-1. 0 means low confidence and 1 is the highest confidence
plates/polygon	Accurate plate polygon
plates/result	Original plate detection result from Snapshot

Error codes

Code	Meaning
400	An error occurred, check response body for more info on the error.

Error Response Schema

{"error":"Invalid Blur intensity. Both faces and plates should not be skipped entirely."}