Two of three · the specimen collection
Reading a century of pressed plants
Museums hold Hawaiian plants collected as far back as 1792, each one photographed and each one showing plainly what it was doing on the day it was picked. Almost none of that has been written down as data. This is the software that reads it — running here, on this page, on real specimens.
8,590
Hawaiian specimens loaded
56
museums
1,918
species
1770–2026
years covered
Step one · running now
Find the parts of the sheet
Press the button. The software fetches a real museum specimen and locates the plant, the label, the colour strip and the packet holding loose seed — reporting how sure it is about each. Then it reads the label. Nothing here is a recording, and none of it is trained: it works from the physics of the image, which is why it runs on museums this project has never seen.
Pick a specimen
BISH 790740 · Pandanus pulposus (Warb.) Martelli · Tuvalu · 1974
Step two · also running now
Train a model on them
Pick a question and a classifier learns from real specimens, then gets tested on ones it has never seen — in about half a second, in this tab, with no server involved. The third question is the one this project actually needs answered, and it is meant to fail. Showing where the method stops is what makes the rest of it worth believing.
What you are about to train
Is this specimen a fern?
Ferns look different from flowering plants at a glance — frond outlines, no flowers, a different way of filling the sheet. A simple model should do well, and if it does not, something is wrong with the setup rather than with the problem.
Split by species, not at random — every sheet of a species goes to one side or the other, so no near-duplicate of a training sheet can appear in the test. Splitting randomly would raise the score and mean nothing.
Step three · what the funding buys
Score what no database records
No museum field separates green fruit from ripe fruit. Neither does any field-observation vocabulary. Yet that is the distinction a control programme turns on.
It is plainly visible in the photograph, and nowhere else. So people with botanical training mark it up, and a model is trained on what they mark. The scoring tool splits fruit into immature, mature and dispersing, hides the collector’s own note until after you commit, and runs on the keyboard — because an expert’s hour is the binding constraint in this whole field.
Try scoring a specimen →The collection
Search all 8,590 Hawaiian specimens →
Served by the museums that hold them — this project stores pointers, never copies · see all 2,100 →
Specimens are searchable by island, elevation, family, museum, and by what the collector recorded about reproduction. A worked example follows one species end to end.