NIST Turns the 600-Page ANSI/NIST-ITL Biometric Data Standard Into a Machine-Readable JSON Specification

The National Institute of Standards and Technology (NIST) has published a machine-readable version of the requirements in ANSI/NIST-ITL 1-2025, the standard that law enforcement and other agencies use to exchange fingerprint, facial and other biometric and forensic data. NIST Internal Report 8616 describes how the agency encoded those requirements as structured JSON data that software tools can process directly, and the tools it built to maintain them.

The standard dates to 1986, when NIST, then the National Bureau of Standards, published a 53-page document focused only on fingerprint identification. It has since added modalities including facial, iris, plantar and DNA data, and the 2025 revision runs to more than 600 pages. NIST says organizations including INTERPOL, the FBI and the Department of Homeland Security rely on it.

Developers and integrators have traditionally read the standard as text and extracted requirements by hand. The report says requirements for a single data element, such as its data type, character limits or number of occurrences, have often been spread and repeated across narrative sections and tables, so each organization had to reconcile them, and separate reconciliations could produce different interpretations.

The machine-readable specification instead represents the standard’s hierarchy of records, fields, subfields and information items in a common JSON structure, checked against a strongly typed JSON schema and a shared library of validation patterns. The data runs to nearly 120,000 lines, so it is kept in partial files that a set of Python scripts merges, validates against the schema and converts into HTML. When validation fails, the scripts point to the specific record and field at fault. NIST built a proof-of-concept HTML version of the standard entirely from the JSON and compared one field, 8.006 in the record type used for signature images, with the original text to confirm the requirements had been captured.

NIST says the specification gives the community a single shared interpretation of the standard and can be used directly by software or adapted into implementation profiles that include only the records and fields a given deployment needs. It is intended for developers, testers and auditors building conformance and data-processing tools, and supplements rather than replaces the text standard.

NIST is releasing the project as a public open-source repository where users can report issues, propose corrections and suggest improvements. Changes to the NIST-maintained version must pass schema validation and subject-matter review before they are incorporated.

Sources: NIST

–

By the ID Tech Editorial Team