CurioAI
Collection digitization

A collection digitization workflow that produces usable records

Digitization succeeds when imaging, metadata, identifiers, quality control, and preservation are planned as one workflow. Producing thousands of disconnected images is not the same as creating a usable digital collection.

01

Choose a purpose and a bounded collection

Define who will use the digital material and what they need to do with it. Access, inventory, conservation, research, insurance, and publication can require different image quality, metadata, rights review, and handling procedures.

Pilot the workflow on a varied sample. Include large, small, reflective, fragile, flat, and poorly documented items so time estimates reflect the actual collection rather than the easiest objects.

02

Connect every image to a stable identifier

Decide the identifier and filename pattern before photography begins. Keep master files, access derivatives, and catalog records linked by that identifier. Avoid filenames based only on camera sequence numbers or staff initials.

Record the photographer, capture date, equipment or method when relevant, rights, and the view represented. These details support accountability and later reuse.

03

Build quality control into each batch

Check focus, exposure, colour, framing, scale placement, file naming, and metadata while the object is still available. Repeating a photograph immediately is much cheaper than retrieving and handling the object again.

A batch should not be marked complete until the expected views, required fields, and file counts reconcile. CurioAI can guide view coverage and draft metadata, while a person confirms the record.

  • Image opens and is sharp at useful magnification
  • Object number matches the catalog record
  • Required views and details are present
  • Colour and scale references are used consistently
  • Rights and sensitivity restrictions are recorded
  • Masters and derivatives follow the storage plan
04

Preserve masters and export the relationship

Keep preservation masters separate from web or working derivatives, with backups in more than one location. Document transformations so future users know which file is original and which is derived.

Export metadata and images together with stable filenames and a manifest. A ZIP package with records, media, and a machine-readable mapping is easier to audit and migrate than a folder of unexplained images.

Before you finish

Review checklist

  • The project has a defined audience and outcome.
  • A representative pilot informs time and cost estimates.
  • Identifiers connect objects, images, and records.
  • Master and access files are distinguished.
  • Quality control happens before objects are returned.
  • Rights and restrictions travel with the record.
  • Backups and export tests are documented.

Where AI fits

Technology can accelerate capture, transcription, field suggestions, and consistency checks. It cannot establish provenance, legal title, cultural authority, or conservation conclusions from an image alone. CurioAI keeps generated fields editable and requires a person to review the record before saving it.

See the guided cataloging workflow
Further reading

Sources and professional guidance

These sources inform the workflow above. Institutions should also follow their own documentation policy, legal requirements, community protocols, and specialist standards.

From guidance to practice

Create a schema for the collection in front of you.

Use photographs, live questions, and an editable extraction plan to build a reviewable catalog record.

Request access