Discover your Data.
Go deeper into your collection.
Natural history collections hold an extraordinary amount of scientific knowledge, but much of it is still locked behind inaccessible handwritten and printed specimen labels. Reading those labels with AI and making them more available to researchers is only the beginning. The real challenge is turning each specimen label into accurate, structured, and connected data. That is the challenge that DEEP by PICTURAE masters
Research-ready records, not blocks of text.
DEEP by PICTURAE reads digitized specimens and delivers research-ready records. Names are checked against the Catalogue of Life and IPNI. Localities are georeferenced and delivered with a map link. Collectors are matched to existing internal lists provided by your institution. Every value shows exactly where on the label it was read.
Read from a real herbarium sheet. The locality was flagged for a curator; the rest passed the confidence threshold.
Dedicated specialists, each with their own role, and room for more.
Labels vary in language, handwriting, layout and age, so DEEP by PICTURAE does not rely on a single AI model. Specialist agents each do one job, and their results are combined into one consistent record. The set is extensible: new specialists and new models can be added as collections and technology evolve.
A plausible answer is not necessarily a correct one.
No matter the type of natural history collection, you set the confidence threshold for your data extraction results. Anything that moves above the established threshold passes into the finished phase automatically, while those that do not will be sent to you directly for review. Your experts will approve or correct the data while checking the records against the original image. Both the original reading and the corrected value are kept, with a complete audit trail.
Interface from the current build.
A collection is a knowledge hierarchy, not a pile of images.
Information known at box or folder level is inherited by every specimen inside it, and conflicts are flagged. Reference data comes built in, and you connect your own collector registers and gazetteers, so the platform draws on existing knowledge rather than asking the AI to guess.
Workflows are configured per project and batch.
In the format your collection system expects.
Workflows are configured per project and batch, and output is delivered with fields aligned to Darwin Core and Dublin Core, or mapped to your own schema.
Ownership, and European infrastructure.
Institutions retain ownership of their images and the data produced from them. Processing takes place on infrastructure operated by PICTURAE, in a region agreed with you, with a self-hosted model option for material that must not be sent to a third-party model provider.
Available today as a service.
No matter who has digitized your specimens, Picturae would love to work with you to process them through DEEP by Picturae. After our system does its work, your curators and experts will review flagged records to ensure the data you receive is exactly how you want it. Your institution gets direct access to our platform, and we are excited to bring this latest development from Picturae to a select few launch partners. Inquire with us at the button below to become a launch partner for DEEP by Picturae!
- Send: your digitized specimens and the fields you need
- Configure: specialists, your lists, your delivery schema
- Run and review: PICTURAE runs the batches; your curators review the flagged sheets
- Deliver: research-ready records in your format
Become a launch partner
DEEP by PICTURAE is offered as a service today. Tell us about your institution and your digitized specimens, and we will get back to you promptly.
