Understanding Veriff's Name Extraction Process and Error Resolution
Veriff employs a robust system to extract and manage names from ID documents. By leveraging the Machine Readable Zone (MRZ) as a primary data source, Veriff ensures accurate name extraction and supports smooth identity verification processes. Below is a comprehensive guide covering the name extraction process, handling of edge cases, error resolution options, and matching rules.
Name Extraction Process
Veriff's name extraction relies on the MRZ, which separates critical name data such as:
Surname: The first name listed in the MRZ.
Given Names: All subsequent names in the MRZ. If multiple names are present, the first is treated as the First Name, and any additional names are considered Middle Names.
For documents with abbreviations (e.g., 'ep.' indicating spouse), Veriff does not extract names following such abbreviations under the 'Surname' field. While there could be potential for classification under an 'Extra names' field, not all integrations currently support this field.
Handling Edge Cases & Abbreviations
Names with Abbreviations: Veriff ignores any name following abbreviations such as 'ep.' for spouses, focusing only on the names explicitly mentioned in the MRZ.
Order of Names: When names on the ID do not follow the typical First Name–Middle Name–Last Name order, Veriff uses MRZ standards to correctly align and separate given names. This ensures a consistent and systematic approach to name extraction.
Multiple Names: If a person has multiple given names, Veriff processes these in alignment with MRZ standards to maintain accuracy in First-Middle-Last name distinctions for verification purposes.
Resolving Name Extraction Errors
If an error occurs during the name extraction process due to issues such as poor image quality or document layout, Veriff provides two primary options for resolution:
Resubmission: Reset the session to allow the end user to re-upload the document with improved image quality for better processing.
Manual Correction: Submit a manual correction request to Veriff Support with the necessary permissions. Veriff's team will rectify the incorrectly extracted data based on the submitted request.
Users can select the appropriate resolution method depending on their specific situation.
Matching Rules and Name Order
Successful verification is often achievable even if the order of names in the ID document differs from the First Name and Last Name submitted in an API request. Veriff considers small differences in name order or the inclusion of middle names and applies flexible matching rules to account for these discrepancies. This ensures that verification can proceed as long as the names refer to the same individual. By relying on MRZ data and comprehensive matching configurations, Veriff accommodates various naming conventions and ensures successful verifications for users worldwide.