What is the Visual Inspection Zone (VIZ) on an ID document?

The Visual Inspection Zone, usually abbreviated VIZ, is the part of an identity document meant to be read directly by a person rather than a machine. It's essentially everything on the document that isn't the MRZ: the printed name, address, date of birth, document number, photograph, signature, and any other visual elements like issuing authority logos or security patterns, all laid out in a way a human can look at and understand without any special equipment.

This is the section a border officer, hotel receptionist, or bank teller has traditionally read to verify someone's identity, long before automated scanning existed. Every identity document has one, and unlike the MRZ, there's no single global standard governing exactly how it should be laid out.

Countries, and often different document generations within the same country, design their visual zones quite differently, with varying fonts, field orders, languages, and visual security features.

That variation is exactly what makes visual zone data extraction a harder technical problem than MRZ extraction. Where MRZ OCR can rely on one fixed font and one fixed field layout defined by ICAO 9303, visual zone OCR has to recognize dozens of different fonts, layouts, and languages, and correctly figure out which piece of text corresponds to which field on each specific document type.

This typically requires a large library of document-specific templates, since the software effectively needs to know, in advance, roughly where the name field sits on a German ID card versus where it sits on a Brazilian passport.

Despite being harder to process automatically, the visual zone carries valuable information the MRZ doesn't include at all: a full address, for one, plus the photo itself, which is essential for any biometric or photo-matching verification step. It's also an independent data source that can be cross-checked against the MRZ, one of the more effective ways of catching data inconsistencies or signs of document tampering.

ScanDoc's document scanning extracts data from both the visual zone and the MRZ, treating them as complementary sources, the visual zone for its broader field coverage and the photo, the MRZ for its speed and built-in check-digit validation, rather than relying on either one alone.