MRZ (Machine Readable Zone) 101
This article provides an overview of the five main types of Machine Readable Travel Documents (MRTDs) as defined by the ICAO Standard. It explains how these MRTDs, along with other Machine Readable Documents, are supported by the Anyline MRZ scanner.
MACHINE READABLE TRAVEL DOCUMENTS
A machine-readable passport (MRP) is a machine-readable travel document (MRTD) with the data on the identity page encoded in optical character recognition format.
These travel documents are standardized by the ICAO Document 9303 and have a special machine-readable zone (MRZ), which is usually at the bottom of the identity page at the beginning of a passport and includes 2 or 3 lines with the OCR-B font text.
TD3 Documents
TD3 sized Machine Readable Travel Documents consist of two lines of 44 characters each.
The full specification can be found in Doc 9303 Part 4 on the ICAO Website.
TD2 Documents
TD2 sized Machine Readable Official Travel Documents consist of two lines of 36 characters each.
The full specification can be found in Doc 9303 Part 6 on the ICAO Website
TD1 Documents
TD1 sized Machine Readable Official Travel Documents consist of two lines of 30 characters each.
The full specification can be found in Doc 9303 Part 5 on the ICAO Website
MRV-A Visa
Format-A Machine Readable Visas consist of two lines of 44 characters each.
The full specification can be found in Doc 9303 Part 7 on the ICAO Website
MRV-B Visa
Format-B Machine Readable Visas consist of two lines of 36 characters each.
The full specification can be found in Doc 9303 Part 7 on the ICAO Website
OTHER MACHINE READABLE DOCUMENTS
MRZ on Swiss Driving License
The MRZ on Swiss driving licenses is not following the ICAO standards. It consists of three lines - one line of 9 and two lines of 30 characters each.
The fields which can be extracted from the MRZ are listed below. (Note that the name of the field in the Anyline result can differ from the name in the explanation image below.) The following list shows the Anyline field name at first and the alternative name in brackets:
Document number
Document type
Issuing country code
Personal number
Optional data
Date of birth
Surname
Given names
MRZ on US Green Card
The MRZ on US Green Cards is not following the ICAO standards. It consists of three lines of 30 characters each and has a special document type. The fields which can be extracted are listed below. Note that the name of the field in the Anyline result can differ from the name in the explanation image below. The following list shows the Anyline field name at first and the alternative name in brackets:
Document type (C1 or C2)
Issuing country code
Document number
Personal number
Date of birth
Sex
Date of Expiry
Nationality
Surname
Given Names
All Check Digits Valid
Since some countries do not fully implement the ICAO Specifications, invalid check digits do not result in an invalid result object. Instead, the allCheckDigitsValid flag is set to false if at least one of the check digits is invalid. To determine the validity of individual check digits, the ICAO check digit algorithm can be found in Doc 9303 Part 3 on the ICAO Website.
If the character at the check digit’s position is not a digit, no valid result will be returned. Exception: If the optional Personal Number field is left empty (only containing ‘<’), an empty Check Digit for the Personal Number (‘<’) is considered valid.
Note - For the MRZ on Swiss driving licenses, this value will always be true. In this custom MRZ, there are no check digits, so the result is always valid with respect to check digits.
Processing of MRZ data with Anyline
To extract the data from MRZ image files obtained by scanning, it must go through the following processing stages:
Step 1 - Preprocessing of scanned images
PLEASE NOTE
You scan an ID document's identity page with MRZ. Environmental conditions and low image quality are a tough factor for robust accuracy with mobile devices. To enhance the input image before the MRZ is extracted some preprocessing is usually applied.
Step 2 - Extracting data from MRZ
Only one MRZ may be captured and processed at a time. The text of each of the 2 or 3 MRZ lines will be recognized and parsed to extract the data fields. Anyline provides the calculation of the checksum of the extracted data according to the ICAO standard. The calculated checksum can be compared with the original checksum, to ensure and deliver an accurate result.
Need help?
Check out the Anyline GitHub repository
Look at our Documentation or contact Anyline Support