From Shoebox
to Searchable.
In One Day.
ScanStationAI pairs a local scanning station with cloud-based AI so small museums, archives, and historical societies can digitize their photograph and postcard collections with a workflow that is practical, teachable, and fast.
Built for the shoeboxes, albums, and postcard drawers — the photographs nobody has had time to scan, and increasingly nobody left to identify.
Built for the places that cannot wait for a full digitization department.
Give volunteers and staff a scanning workflow they can learn in an afternoon, while the system handles description drafts, tagging, and structured organization in the background.
Small institutions do not just need a scanner. They need a workflow that keeps a collection in order, captures context while people still remember it, and turns a box of prints into something searchable and useful.
Too much is boxed up
Irreplaceable photographs, postcards, and albums stay hidden because digitization projects feel too big to start.
Vendors are expensive
Traditional outsourcing works for large institutions, but not for the shoebox, drawer, and room-scale projects most local museums face.
Nobody knows who is in the picture
A scan of an unlabelled print is still an unlabelled print. Photographs need dates, names, places, and subjects attached before the people who remember them are gone.
Knowledge is walking out the door
Volunteers and staff know who donated the album, who is in the photo, and why it matters. That context disappears if it is not captured now.
How ScanStationAI Works
A local capture station handles the physical workflow. The cloud handles description, tagging, and publishing-ready metadata.
Local scan station
High-speed feeding for loose prints and postcards — roughly a second each — with overhead capture for bound albums and anything too fragile to feed.
Operator-guided organization
Boxes, albums, and item order stay visible while the operator works, so the digital record mirrors the physical one.
AI-assisted description
Titles, descriptions, subjects, and tags are suggested automatically, so a shoebox comes out the other end searchable instead of merely scanned.
Both sides, one pass
Fronts and backs scan together, so the caption someone pencilled on the reverse in 1954 stays attached to the image it explains.
What the Operator Actually Does
The operator stays in control. The system is there to accelerate repetitive work, not replace collection knowledge.
Not another locked-in silo.
ScanStationAI is being built to fit into the collection systems museums already use, and the standards they already care about.
IIIF support
Level 2 IIIF support is part of the plan so collections can move toward interoperable image delivery and presentation.
Dublin Core metadata
Item records are stored in Dublin Core terms rather than a private schema, so descriptions move in and out of other systems without being translated first.
Persistent ARK identifiers
Every item gets an ARK: a permanent identifier registered under our own NAAN, so a link to a photograph keeps resolving even as websites and systems change.
Import paths
We already import from PastPerfect and Omeka / Omeka S. Contact us for a demo of your public data in our system.
Coming soon
A fuller CMS inspired by Omeka S RDF-style internal structure, plus authority list integration for cleaner names, places, and subjects.
Want to see it in action?
We are looking for organizations to try this with and tell us where it falls short. If you want to talk about a pilot, an in-person demonstration, or a future grant-backed project, reach out.