MRZ in Passports: What It Is and How Passport Verification Works

When we look at a passport, we usually focus on the photo, name, passport number, nationality, and expiry date. But at the bottom of the passport data page, there is a block of text that looks difficult to read for humans but is very important for machines. This area is called the Machine Readable Zone, commonly known as the MRZ.

The MRZ is one of the most important parts of modern passport verification. It allows border control systems, airport kiosks, airline check-in counters, hotel registration systems, and identity verification platforms to quickly extract passport data in a standardized format.

However, MRZ verification is often misunderstood. Reading the MRZ is not the same as proving that the passport is genuine. It is only one layer in a complete document verification process.

What Is the MRZ?

The MRZ is the machine-readable text printed at the bottom of a passport’s biographical data page. In most passports, it appears as two lines, and each line contains 44 characters.

A typical passport MRZ looks like this:

P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<
L898902C36UTO7408122F3404159ZE184226B<<<<<10

This may look like random text, but it follows a strict international structure. Each character position has a meaning. The MRZ stores key passport and identity information such as:

• Document type
• Issuing country
• Surname and given names
• Passport number
• Nationality
• Date of birth
• Sex
• Expiry date
• Optional personal number or document-specific data
• Check digits

The < symbol is used as a filler character. It replaces spaces, separates name parts, and fills unused positions.

Why Is the MRZ Important?

The MRZ was designed to make travel documents easier and faster to read by machines. Before machine-readable passports, officers had to manually read and enter passport data, which was slow and error-prone.

With MRZ, a system can scan the passport and extract structured data almost instantly. This improves:

• Speed at border control
• Accuracy of data entry
• Airline check-in efficiency
• Automated gate processing
• Watchlist and database checks
• Digital identity verification workflows

In simple terms, the MRZ converts the passport data page into a predictable machine-readable format.

Structure of a Passport MRZ

A passport normally uses the TD3 format, which contains two MRZ lines of 44 characters each.

Line 1: Document Type, Issuing Country, and Name

The first line contains the document type, issuing country, and the holder’s name.

Example:

P<UTOERIKSSON<<ANNA<MARIA<<<<<<<<<<<<<<<<<<<

This can be understood as:

• P< — document type, where P indicates passport
• UTO — issuing country or organization code
• ERIKSSON<<ANNA<MARIA — surname and given names
• < — filler characters to complete the fixed line length

In MRZ naming rules, the surname and given names are separated by <<. Spaces inside names are represented by a single <.

For example:

ERIKSSON<<ANNA<MARIA

Means:

Surname: ERIKSSON
Given names: ANNA MARIA

Line 2: Passport Number, Dates, Nationality, and Check Digits

The second line contains document and personal data.

Example:

L898902C36UTO7408122F3404159ZE184226B<<<<<10

This line includes:

• Passport number
• Passport number check digit
• Nationality
• Date of birth
• Date of birth check digit
• Sex
• Expiry date
• Expiry date check digit
• Optional data
• Optional data check digit
• Composite check digit

This line is very important during verification because it contains multiple check digits used to detect OCR errors, manual entry errors, or possible data manipulation.

What Are MRZ Check Digits?

A check digit is a single digit calculated from a specific data field. It helps confirm whether the field was read correctly.

For example, the passport number has a check digit. The date of birth has a check digit. The expiry date has a check digit. There is also a final composite check digit that validates multiple fields together.

The calculation uses a repeating weight pattern:

7, 3, 1, 7, 3, 1...

Characters are converted into values:

• Numbers 0–9 keep their numeric value
• Letters A–Z are converted to values 10–35
• The filler character < has value 0

Each character value is multiplied by the weight at its position. The results are added together, and the remainder after division by 10 becomes the check digit.

For example, if a system reads one character incorrectly, the check digit will usually fail. This is useful for detecting OCR mistakes such as:

• O read as 0
• I read as 1
• B read as 8
• missing or extra filler characters
• poor image quality
• glare or blur on the passport page

MRZ Verification Process

A good passport verification process does not simply read the MRZ and accept it. It should follow multiple verification layers.

1. MRZ Detection

First, the system must locate the MRZ area on the passport image. This can be done using document detection, image processing, OCR, or a trained computer vision model.

Good image capture is very important. The system should check for:

• Clear focus
• Proper lighting
• No glare
• Full passport page visible
• Correct orientation
• No cropping of the MRZ area
• Sufficient resolution

If the MRZ is blurred or partially hidden, even a good OCR engine may produce incorrect results.

2. OCR Reading

After detecting the MRZ, OCR is used to read the two lines. MRZ uses a standardized font and character set to improve machine readability.

The OCR result should not be trusted blindly. It must be validated against expected MRZ rules.

For a passport MRZ, the system should confirm:

• There are exactly two lines
• Each line has exactly 44 characters
• Only valid MRZ characters are present
• Field positions match the expected TD3 format
• Filler characters are used correctly

3. Field Parsing

Once the MRZ text is extracted, the system parses each fixed field.

For example:

• Characters 1–2 of line 1: document code
• Characters 3–5 of line 1: issuing country
• Characters 6–44 of line 1: name
• Characters 1–9 of line 2: passport number
• Characters 11–13 of line 2: nationality
• Characters 14–19 of line 2: date of birth
• Characters 22–27 of line 2: expiry date

Because MRZ uses fixed positions, parsing should be deterministic. A reliable parser should not guess field boundaries.

4. Check Digit Validation

Next, the system recalculates all check digits and compares them with the digits printed in the MRZ.

This step can detect many common reading errors.

For example:

• If the passport number is read incorrectly, the passport number check digit may fail.
• If the date of birth is misread, the date of birth check digit may fail.
• If multiple fields are altered, the composite check digit may fail.

A failed check digit does not always mean fraud. It may also mean poor OCR quality. The system should handle this carefully and return a clear reason such as:

MRZ check digit validation failed. Please recapture the passport image clearly.

5. MRZ vs VIZ Comparison

The Visual Inspection Zone, or VIZ, is the human-readable area of the passport data page. It contains the printed name, passport number, nationality, date of birth, sex, and expiry date.

A strong verification system should compare MRZ data with VIZ data.

For example:

• MRZ passport number should match the printed passport number
• MRZ date of birth should match the printed date of birth
• MRZ expiry date should match the printed expiry date
• MRZ name should reasonably match the printed name

This comparison is important because a forged document may contain inconsistent data between the visual area and the MRZ.

However, name comparison must be handled intelligently. MRZ removes punctuation, changes spaces to filler characters, may truncate long names, and may not represent special characters exactly as shown in the VIZ.

6. Document Authenticity Checks

MRZ validation confirms that the MRZ is structurally correct. It does not prove that the passport is genuine.

A fake passport can still contain a valid-looking MRZ with correct check digits. Therefore, additional document authenticity checks are required.

These may include:

• Template validation
• Font and layout checks
• Security feature inspection
• UV/IR image analysis
• Hologram and laminate checks
• Photo substitution detection
• Data page tampering detection
• Chip verification for ePassports

In professional border control or identity verification systems, MRZ is only one part of a larger decision engine.

MRZ and ePassport Chip Verification

Modern biometric passports usually contain a contactless chip. The chip stores electronic data such as the holder’s facial image and document data.

In ePassport verification, the MRZ can be used as an input to access the chip, depending on the security mechanism supported by the document. Once the chip is read, the system can compare:

• MRZ data vs chip data
• VIZ data vs chip data
• Printed portrait vs chip portrait
• Live face vs chip portrait

The strongest verification practice includes cryptographic validation of the chip data. This helps confirm that the data was issued by a trusted authority and has not been modified.

This is why chip verification is much stronger than MRZ-only verification.

Common MRZ Verification Failures

In real-world systems, MRZ verification may fail for many reasons.

Poor Image Quality

Blur, glare, low resolution, shadows, or reflections can cause OCR errors.

Cropped Passport Page

If the bottom part of the passport is missing, the MRZ cannot be read completely.

Wrong Document Type

Some systems expect a passport MRZ but receive an ID card or visa MRZ. These formats are different.

OCR Confusion

Characters such as 0, O, 1, I, 5, S, 8, and B may be confused.

Expired Passport

The MRZ may be valid, but the expiry date may show that the document is no longer valid.

MRZ and VIZ Mismatch

The MRZ may not match the human-readable printed data, which can indicate OCR failure, poor parsing, or document manipulation.

Unsupported Country or Format Variation

Although ICAO standards define the structure, real-world passports may still have country-specific handling requirements, especially in optional data fields.

Best Practices for MRZ Verification Systems

A reliable MRZ verification system should follow these best practices:

1. Validate image quality before OCR.
2. Detect and correct document orientation.
3. Ensure the full MRZ area is visible.
4. Enforce exact MRZ length and character rules.
5. Parse fields using fixed positions, not guessing.
6. Recalculate all check digits.
7. Compare MRZ data with VIZ data.
8. Check document expiry.
9. Use country-code validation.
10. Perform chip verification when available.
11. Return clear error messages to the user.
12. Log verification failure reasons for audit and improvement.

For example, instead of showing a generic error like:

Verification failed.

A better system should show:

The MRZ could not be verified because the passport image is blurred. Please retake the photo with the full passport page clearly visible.

Clear feedback improves user experience and reduces support cases.

MRZ Verification Is Not Enough

The most important point is this: MRZ verification is a data consistency check, not full document authentication.

MRZ can tell us whether the machine-readable data is formatted correctly and whether its check digits are valid. It can also help detect OCR errors and simple inconsistencies.

But MRZ alone cannot confirm:

• Whether the passport is physically genuine
• Whether the document was issued by a trusted authority
• Whether the photo was replaced
• Whether the data page was tampered with
• Whether the passport holder is the rightful owner

For high-security use cases, MRZ verification should be combined with document authenticity checks, chip verification, biometric face matching, and backend watchlist or database checks.

Conclusion

The MRZ is a small but powerful part of the passport. It allows machines to read passport data quickly, accurately, and consistently. It supports faster border processing, automated identity verification, and reliable data extraction.

However, MRZ verification should be treated as one layer in a complete passport verification process. A valid MRZ is a good sign, but it is not final proof of authenticity.

The best verification systems combine MRZ parsing, check digit validation, VIZ comparison, document security analysis, chip verification, and biometric matching. This layered approach provides stronger protection against OCR errors, document tampering, and identity fraud.

In passport verification, the MRZ is the starting point — not the final decision.