Entities Object
The entities
field is a property of the document_page_object and contains both extracted and not extracted fields,
defined in the service_fields
, from the processed document page.
Note
aggregated_fields are not part of this object.
Each property within entities
represents the identifier of one of the defined service_fields,
and the associated value can be either a list of Entity
objects or a single Entity
object,
depending on whether the field has multiple values or not.
Note
The properties value type (object
or list[object]
) are fixed and will never change.
Example:
if name_1
is configured as multiple_values: true
you can always consider name_1
type as list[object]
.
Entity Object
Schema
Schema: Entity Object
Key | Type | Description |
---|---|---|
id | integer | a page unique ID assigned to the entity |
text | string | the extracted text |
confidence | float | the confidence of the prediction |
bounding_boxes | list | list of polygon objects |
Example of entities
entities example with single and multiple values
{
"entities": {
"surname_1": [
{
"id": 0,
"text": "VERDI",
"confidence": 99.43,
"bounding_boxes": [
[
[
290,
287
],
[
369,
287
],
[
369,
306
],
[
290,
306
]
]
]
}
],
"name_1": [
{
"id": 1,
"text": "GIUSEPPE",
"confidence": 99.47,
"bounding_boxes": [
[
[
291,
323
],
[
422,
323
],
[
422,
341
],
[
291,
341
]
]
]
}
],
"fiscal_code_1": {
"id": 2,
"text": "VRDGPP13R10B293P",
"confidence": 90.97,
"bounding_boxes": [
[
[
268,
236
],
[
543,
236
],
[
543,
255
],
[
268,
255
]
]
]
}
}
}