Skip to content

File 0200Updated 2026-07-26Owns textile verification software

Textile verification software, built as an evidence layer.

Four surfaces, one record. TexFile reads the documentation a garment already carries, reconciles it into a verified item record with calibrated confidence, and serves that record to whoever needs to prove something: the sorting line routing the material, the brand substantiating a claim, the scheme reporting a mass flow.

§Surfaces

What a verification and evidence layer is

Most software in textile circularity moves claims around. A traceability platform distributes what a brand already holds. A certificate records what one party asserted on one day. Neither one originates evidence, and neither one can say how sure it is.

An evidence layer sits underneath all of that. It reads primary documents, reconciles them against each other, and writes a record in which every value names the source it came from and carries a score for how well the sources agreed. Everything above it, from a routing decision on a sorting line to a Digital Product Passport export, is then derived from one auditable object rather than assembled from assertions.

The three surfaces of the product

What the system does
Record building
A multi view capture station photographs each item. Vision models read composition, care, origin and brand from labels in any language or condition, and the raw text is normalized into controlled EU vocabularies with a confidence on every field.
Record store and API
Every record is stored with its provenance and served through an API built for the line: a routing answer inside a 50 ms ceiling, and the full record for anyone who needs to inspect how the answer was reached.
Exports and reporting
The same record is emitted as a Digital Product Passport payload, as chain of custody events, or as mass flow evidence per fraction. One reconciliation, several obligations discharged.

The four pages below are those surfaces seen from where each buyer stands. They describe one system for used textiles, not four products.

Who it is built for

Four buyer tracks

  • Sorting and recycling operators

    To know what an item is while it is still on the belt, and to prove later why it was routed the way it was.

    A routing answer per item inside the line's latency budget, and a persistent record of the decision and its evidence.

  • Brands and retailers

    To substantiate recycled content and prepare for the Digital Product Passport, including for stock that carries no digital identifier.

    Verified recycled content and chain of custody records for legacy garments, exportable in the passport format.

  • Producer responsibility organisations

    To evidence what was collected, sorted and recycled, and to justify eco modulated fees rather than assert them.

    Auditable mass flow evidence per fraction, aligned with the reporting standards the schemes already use.

  • Public waste authorities

    To account for separately collected volumes under KrWG §20 with numbers that survive an audit.

    Per fraction evidence from the sorting line itself, rather than an estimate reconstructed afterwards.

What every record conforms to

Built to conform to
EU Digital Product Passport
The record is assembled against the passport data model, so an export is a projection rather than a migration.
GS1 identifiers and Digital Link
Stable identifiers, resolvable the way the rest of the supply chain already resolves them.
EPCIS
Custody events in the standard vocabulary, so a chain of custody reads the same to every party in it.
SHACL
Shape validation before a record is sealed. A record that does not conform does not ship.
PROV-O
Provenance as a first class property of the data, not as a note in a comment field.

Common questions

What is textile verification software?

Software that establishes what a textile item is and where the claim comes from, rather than storing a claim somebody else made. It reads primary documents such as care and composition labels, reconciles them against each other, and produces a record in which every field names its source and carries a calibrated confidence.

How is this different from a traceability platform?

Traceability platforms distribute data that brands already hold about new, tagged products. TexFile originates data for items that carry no digital identifier, at the point where they are sorted, and feeds those platforms through open standards. It is the layer beneath them rather than an alternative to them.

Does it replace NIR sorting hardware?

No. Near infrared sensing reads the dominant surface fibre and returns a material signal. TexFile reads the declared blend, care, origin and brand from the labels and returns a record. The two are complementary, and TexFile runs as a software retrofit on cameras a plant already operates.

What does calibrated confidence mean in practice?

Each extracted value carries a score, and the score means what it says: at a 0.9 threshold, nine in ten such values hold. The operator sets the threshold rather than the vendor. Items above it route automatically, items below it go to assisted review, and every correction is written back with its provenance.

Verify with us

See a record traced end to end.

The fastest way to judge an evidence layer is to follow one field back to the document it came from. Ask for a walkthrough on a record.

Book a demo