Overview

This node uses Platerecognizer's Snapshot API to read license plates from vehicles that are detected using the Vehicle & License Plate models.

This node requires AI Model Node (Vehicle Detection model) -> Model Inference Node (License Plate model) -> Track Objects Node prior to it in order to function properly using the "LPR & Make/Model/Color" recognition mode.

Inputs & Outputs

Inputs : 1, Media Format : Raw Video
Outputs : 1, Media Format: Raw Video
Output Metadata: license-plate results, counts, and entered/exited deltas under nodes.<node_id>, plus associated object attributes.

Properties

Property	Description	Type	Default	Required
`pr_mode`	Specify which platerecognizer api to use. Options: Snapshot cloud (`snapshot_cloud`): API hosted by Platerecognizer; Snapshot On-premise (`snapshot_onprem`): API hosted by you using Platerecognizer docker SDK.	`enum`	`snapshot_cloud`	No
`pr_api_key`	Snapshot API token from here. Conditional on `pr_mode` being `snapshot_cloud`.	`string`	`null`	Yes
`pr_host`	Base URL for your Platerecognizer On-premise installation. See here for instructions. Conditional on `pr_mode` being `snapshot_onprem`.	`string`	`http://localhost:8080`	Yes
`geographies`	Geographies to consider for license plate recognition. Leave blank to consider all. Region codes are available here.	`string`	`us`	No
`objects_to_track`	ex. `car`, `vehicle`, `truck`. Classes to consider on the node. Leave blank to consider all.	model-labels	`car,vehicle,truck`	No
`display_info`	Display LPR info on video?	`bool`	`true`	No
`recognition_mode`	Attempt recognition until the above info is found. Options: LPR & Make/Model/Color (`lpr_mmc`); LPR Only (`lpr_only`); Make/Model/Color Only (`mmc_only`).	`enum`	`lpr_mmc`	No
`requests_buffer`	Time that the object must be present before the initial LPR request and between subsequent retries. Unit: seconds.	`float`	`1.0`	Yes
`presence_buffer`	Time during which objects that leave and re-enter the scene will be ignored. Increase to reduce duplicate alerts for a given object. Unit: seconds.	`float`	`5.0`	Yes
`min_dims`	Don't lookup objects smaller than this size. Unit: pixels.	`number`	`20`	No
`max_attempts`	Only attempt to lookup an object these many times before giving up.	`number`	`5`	No

Output Metadata

The fields below are declared by this node's metadata schema; the JSON values are representative examples.

Path	Type	Description
`nodes.<node_id>.license_plates_entered_delta`	`integer`	Number of new license plates detected since last frame
`nodes.<node_id>.license_plates_exited_delta`	`integer`	Number of license plates that left since last frame
`nodes.<node_id>.license_plates_count`	`integer`	Total number of license plates in the view right now
`nodes.<node_id>.license_plates.<object_id>.plate.type`	`string`	Recognition result category for the plate.
`nodes.<node_id>.license_plates.<object_id>.plate.score`	`number`	Overall plate-recognition confidence.
`nodes.<node_id>.license_plates.<object_id>.plate.props.plate[].value`	`string`	Recognized plate text.
`nodes.<node_id>.license_plates.<object_id>.plate.props.plate[].score`	`number`	Confidence of this plate-text candidate.
`nodes.<node_id>.license_plates.<object_id>.plate.props.region[].value`	`string`	Recognized plate region.
`nodes.<node_id>.license_plates.<object_id>.plate.props.region[].score`	`number`	Confidence of this region candidate.
`nodes.<node_id>.license_plates.<object_id>.vehicle.type`	`string`	Recognized vehicle type.
`nodes.<node_id>.license_plates.<object_id>.vehicle.score`	`number`	Overall vehicle-recognition confidence.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.make_model[].make`	`string`	Recognized vehicle make.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.make_model[].model`	`string`	Recognized vehicle model.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.make_model[].score`	`number`	Confidence of this make/model candidate.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.orientation[].value`	`string`	Recognized vehicle orientation.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.orientation[].score`	`number`	Confidence of this orientation candidate.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.color[].value`	`string`	Recognized vehicle color.
`nodes.<node_id>.license_plates.<object_id>.vehicle.props.color[].score`	`number`	Confidence of this color candidate.
`nodes.<node_id>.error`	`string`	Most recent processing or recognition error, if any.
`nodes.<node_id>.type`	`string`	Identifies the node type that produced this metadata.

JSON example

{
  "nodes": {
    "lpr_platerecognizer1": {
      "error": "value",
      "license_plates": {
        "object_id": {
          "plate": {
            "props": {
              "plate": [
                {
                  "score": 0.0,
                  "value": "value"
                }
              ],
              "region": [
                {
                  "score": 0.0,
                  "value": "value"
                }
              ]
            },
            "score": 0.0,
            "type": "value"
          },
          "vehicle": {
            "props": {
              "color": [
                {
                  "score": 0.0,
                  "value": "value"
                }
              ],
              "make_model": [
                {
                  "make": "value",
                  "model": "value",
                  "score": 0.0
                }
              ],
              "orientation": [
                {
                  "score": 0.0,
                  "value": "value"
                }
              ]
            },
            "score": 0.0,
            "type": "value"
          }
        }
      },
      "license_plates_count": 0,
      "license_plates_entered_delta": 0,
      "license_plates_exited_delta": 0,
      "type": "lpr_platerecognizer"
    }
  }
}

Object labels and attributes

Object labels/classes added: None.
Object attribute labels/classes added: Recognized plate text (10500), vehicle make (10501), model (10502), color (10503), and region (10504).

The following information is added to the detected object's "attributes" array:

"class_id" field	"label" field	"probability" field
10500	The recognized license plate (string)	LP recognition confidence
10501	Vehicle's make	LP recognition confidence
10502	Vehicle's model	LP recognition confidence
10503	Vehicle's color	LP recognition confidence
10504	Plate region	LP recognition confidence