Overview

---

Intelligent Document Processing (IDP)

This API allows you to extract information from documents using Optical Character Recognition and advanced machine learning NLP algorithms.

Prerequisites

A private IDP service A private URL provided by NeuralSpace to access the IDP service

Endpoint

https://<YOUR-IDP-SERVICE-URL>/ocr/reader

Request

Parameters

The following parameters can be used in the request:

Parameter	Type	Required	Description
image	file	Yes	The image file to be processed.
pageLimit	string	Yes	The maximum number of pages to process.
documentType	string	Yes	Currently only passport, id_card, id_card_old, trade_license, trade_license_detailed documentType are supported

Example Request

The following example shows how to make a request to the API using curl:

curl --location 'https://<YOUR-IDP-SERVICE-URL>/reader' \
--form 'file=@"YOUR_FILE_PATH"' \
--form 'pageLimit="2"' \
--form 'documentType="passport"'

Response

Parameters

The following parameters can be returned in the response:

Parameter	Type	Description
entities	object	The extracted entities and their values from the uploaded document.
texts	array	An array of strings containing the extracted text from uploaded document.

Example Response

The following example shows the response to the above request:

{
    "entities": {
        "name": "SITA MAHA LAKSHMI",
        "surname": "RAMADUGULA",
        "document_number": "JB369854",
        "birth_date": "590923",
        "expiry_date": "211010",
        "sex": "F",
        "nationality": "IND",
        "document_type": "P"
    },
    "texts": [
        "WINS TUIRTGW REPUBLIC OF INDIA",
        "",
        "P IND JE369854",
        "RAMADUGULA",
        "",
        "SITA MANA LAKSHNI",
        "",
        "INDIAN F 2370971959",
        "",
        "GUNDUGOLANU",
        "HYDERABAD",
        "R 8 Lokam",
        "",
        "117107201 1071072021",
        "",
        "P<INDRAMADUGULA<<SITA<MAMA<LAKSHMI<<<<<<<<<<",
        "JB369854<4IND5909234F2110101<<<<<<<<<<<<<<<8",
        ""
    ]
}

Errors

The following error messages can be returned by the API:

Error Message	Description
InvalidImageFormat	The image file provided is not in a supported format.
InvalidImageSize	The image file provided exceeds the maximum allowed size.
InvalidPageLimit	The pageLimit parameter provided is not valid.
InvalidDocumentType	The documentType parameter provided is not valid.
RateLimitExceeded	The requests for the poc server is limited to 30 requests/minute

If an error occurs, the response will have the following format:

{
    "status": "error",
    "message": "<error message>"
}

• Currently id_cards_old documentType used for Old Emirates ID accepts only pdf with both front and back side.
• For all other document types both pdf and image formats are supported
• Accuracy in the test environment may be slightly lower than the test environment