Vehicle Make and Model Recognition Results
The following XML shows a single record produced by vehicle make and model recognition.
<record>
...
<trackname>VehicleMakeModel.Result</trackname>
<VehicleModelResult>
<id>e2fb50f1-a71a-46e0-8af6-c5e7e5ba3afc</id>
<grillepatch>
<point>
<x>379</x>
<y>680</y>
</point>
<point>
<x>739</x>
<y>680</y>
</point>
<point>
<x>739</x>
<y>784</y>
</point>
<point>
<x>379</x>
<y>784</y>
</point>
</grillepatch>
<identity>
<identifier>FocusRS</identifier>
<database>Cars</database>
<confidence>77</confidence>
<metadata>
<item>
<key>type</key>
<value>hatchback</value>
</item>
</metadata>
</identity>
<grillepercentage>100</grillepercentage>
<vehiclemake>Ford</vehiclemake>
<vehiclemakescore>100</vehiclemakescore>
<region>
<left>296</left>
<top>496</top>
<width>501</width>
<height>416</height>
</region>
<regionscore>100</regionscore>
<vehicletype>car</vehicletype>
<vehicletypescore>98</vehicletypescore>
</VehicleModelResult>
</record>
The record contains the following information:
id- a unique identifier for the recognized number plate. The number plate recognition engine issues an ID for each detected appearance of a number plate, and the records output from the vehicle make/model recognition engine use the same IDs as the input records.grillepatch- the region of the input image that was used for recognizing the vehicle make and model.-
identity- represents a match between the ingested media and a vehicle model in your training database. This element contains data only if you train Media Server to recognize vehicle models, you set theDatabaseparameter when you configure the task, and a match is found. The following information is available:identifier- the identifier of the recognized vehicle model.database- the name of the database in which the vehicle model exists.confidence- the confidence score for the match (from 0 to 100, where 100 is maximum confidence).metadata- the metadata that you associated with the vehicle model when you trained Media Server. If there is no metadata in the training database, this element is omitted.
grillepercentage- the percentage of the vehicle grille that was visible in the source image.vehiclemake- the recognized vehicle make.vehiclemakescore- the confidence score for vehicle make recognition, from 0 to 100, where 100 indicates maximum confidence.region- a bounding box that contains the entire vehicle.regionscore- the confidence score for the region (from 0 to 100, where 100 is maximum confidence).vehicletype- the vehicle type (bus,car,motorcycle, ortruck).vehicletypescore- the confidence score for the vehicle type (from 0 to 100, where 100 is maximum confidence).