Passports and ID documents — read in seconds.

ID and passport data extraction is the process of turning a photo or scan of a government-issued identity document into structured fields a KYC, onboarding, or background-check workflow can ingest. The work that used to take a verification analyst a minute or two per ID — read each field, type it into the right form, double-check the format — runs in a few hundred milliseconds and never typos a passport number.

On the sample passport above, Ztract returns the holder's surname, given names, and sex; the date of birth, place of birth, and nationality; the passport number with its date of issue and date of expiration; the issuing authority; and any printed endorsements. The sample is a US passport whose holder was born in Egypt — place of birth and nationality are kept as distinct fields rather than collapsed into one.

The schema generalizes across document types. Passports, national identity cards, driver's licenses, and residence permits all map to the same shape — names, dates, document number, issuing authority — and each field comes back in a normalized form (dates as ISO-parseable strings, names preserved in their printed script with Latin transliteration when both appear). Glare, holograms, and security-watermark overlays don't obscure the underlying values.