What information is encoded in the MRZ?

The MRZ packs a surprising amount of structured personal and document data into a small space, using fixed character positions so a reader always knows what each part means. The exact fields depend on the document type, but a passport's MRZ, the TD3 format, typically includes: document type code, issuing country or organization, the holder's surname and given names separated by <<, passport number, nationality, date of birth, sex, expiry date, and an optional personal number field such as a national ID number.

Interspersed among these are check digits: single numbers calculated from surrounding data that let software confirm nothing has been altered.

ID cards using the three-line TD1 format encode similar fields but split slightly differently across the lines, often including the document number, date of birth, sex, expiry date, nationality, and name, again with check digits woven throughout. A composite check digit at the end of the MRZ validates the whole block together, not just individual fields, which makes it harder to tamper with one piece of data without the whole thing failing validation.

What's notably absent is anything visual. There's no photo, no signature, no address in most cases.

The MRZ is strictly the compact, standardized identity data needed to confirm who someone is and whether their document is still valid, not the full personal profile you'd find in the visual zone. That's a deliberate design choice.

It keeps the format short, consistent, and fast to scan, while relying on the visual zone, and for e-passports, the chip, to hold the rest.

For a business using document scanning, this matters practically. If your onboarding flow needs an address, MRZ data alone won't get you there.

You'd need visual zone OCR or, for e-passports, the RFID chip. ScanDoc's data extraction combines MRZ output with visual zone OCR precisely because the two sources are complementary.

MRZ gives you fast, standardized core fields with built-in tamper checks, and the visual zone fills in everything else while providing a second data point to cross-validate against.