1 · Upload plant photo(s)
Click to choose or drag & drop
Up to 5 images of the same plant · JPG/PNGThe photo is sent to your local demo server, which forwards it to the Pl@ntNet API using your secret key. The key is never exposed to the browser.
2 · Request & response
Upload a photo and click Identify plant to see the request sent to Pl@ntNet and the identification results.
ℹ️ About this demo & the path to production
What this demo shows: live plant identification against Pl@ntNet's global model — the real request and response an officer would evaluate for accuracy and fit.
🪷 Local / vernacular names (Hindi & Chhattisgarhi)
Pl@ntNet returns scientific names and mostly English common names — it does not provide Hindi/Chhattisgarhi names. In production we would maintain a species → vernacular-name master table (keyed by scientific name / GBIF ID) curated with the CG Forest Dept., and display the local name alongside every result. This demo shows the identification layer; the local-name mapping is a thin lookup we add on top.
🌏 CG / India-specific accuracy
This uses the world flora model. For higher accuracy on CG's native & medicinal flora, options are: restrict to a regional Pl@ntNet flora, or train a custom model on a CG-labelled image dataset (with local-name labels built in).
🛠️ Productionisation
Secrets management for the key, request logging & audit, rate-limiting, offline/low-bandwidth field capture, a species knowledge card (uses, toxicity, habitat), and multi-image capture guidance for field staff.
Avenoir · Pl@ntNet API demo. Identification results are provided by Pl@ntNet (my.plantnet.org) and are advisory.