GBG Developers

Guides   |   Verify Document Image

Verify Document Image

Our Verify Document Image service gives you the ability to enhance the experience of your users by accepting images of documents and using them to perform validations as part of your collection of verification checks.

What does verify document image do?

Verify document image allows you to submit images of the identifying documents of your end users, check them for tampering and other fraudulent activities and extract the data contained within them.

For a list of supported documents please follow this link and review the latest library release.

Multi-part documents that contain different sides / pages can be seen by the addition of a (Front) and (Back) suffix on a document type.  At the moment only the 'German National Identification (Front)' and 'German National Identification (Back)' have functionality prompt for the other side via the 'action' field and once submitted to cross match details from both sides to ensure they are from a single document.

How does verify document image work?

The Request

The verify document service allows you to specify a level of tolerance in the 'tolerance' header of the request. The tolerance options are 'DEFAULT', 'MEDIUM', and 'STRICT'. The stricter you are with the tolerance level the more likely a document is to be rejected by the service.

Images are submitted to the service using a POST request, supplying your image as multipart/form-data.

The Person Id

If this is the first request you are making in a series of verification requests that relate to a single user, you do not need to supply a person-id, and one will be created for you by IDAAS and returned in the response headers with the key 'person-id'. You should use this id for each subsequent request you make to any of the verification services. This will allow IDAAS to combine these verifications into a single record and perform actions such as combined decision

If you have already performed other verification checks and already have a person-id, this should be supplied in the URL. For more information on where and how to supply a person-id, see our API Reference Documentation.

The Response

Identifiers for this specific verification request, and your submitted documents are returned in response headers, named 'person-id', 'verification-id' and 'document-id'. These can be used with the Verification History UI to retrieve your submitted requests at a later date. Your 'verification-id' will be required for performing further checks that rely on the document such as Facematch. The 'document-id' is used to submit multi-part document sides (e.g. submitting a German National Identification Card (Front) and then subsequently submitting the (Back) of the same document).  The address returned in a Verify Document response will have been cleansed by default, so subsequent requests (such as Verify Person) should set the cleanse-address header to false.  Alternatively the Verify Document cleanse-address can be turned off, but this should be done with caution because reading text from a document image can result in misreading and strange results (as it is very dependent on lighting conditions).

The API will return fields that have been extracted (read from) the document image.  There will always be 'raw' fields returned for all supported document types in the library.  There are two document types that will be normalised so that their 'raw' fields are structured for subsequent Document Data calls, these are:

  • UK Driving Licence
  • German European ID

A successful response can be see in our API Reference Documentation.

The verification result

The possible values of the verify document image result are as follows:

  • PASS:Passed
  • Failed
  • Undetermined

What next?

To implement the verify document image API, find detailed API information in our API Reference.