← Wiki

Ingredient

iNaturalist API

Also known as: iNat API, iNaturalist, Seek by iNaturalist

Open API to the iNaturalist citizen-science database — 200+ million species observations from millions of contributors worldwide, with photographs, geolocation, and expert-verified taxonomic IDs. The right ingredient for any farm-tech project that needs species identification (weeds, pests, beneficials, plants in a forage stand), regional species checklists, or a labeled-image source for training custom CV models. The companion *Computer Vision API* runs iNaturalist's own trained model — submit a photo, get a ranked list of species suggestions. Free for non-commercial use; rate-limited (60 req/min default).

Inputs / outputs

  • Endpoints:
    • /v1/observations — search and retrieve observations by location, taxon, date
    • /v1/computervision/score_image — submit photo, receive ranked species
    • /v1/taxa — taxonomic lookup
    • /v1/places — geographic regions
  • Output: JSON
  • Rate limit: 60 req/min default; higher for partners
  • Auth: OAuth2 for write operations; read endpoints require no auth

Solves / unlocks

  • Field species identification from a captured photo (no local ML)
  • Bioregional species checklists (the basis for the bioregion-intake.py script in this vault)
  • Labeled-image training corpus (download verified-research-grade observations as training data)
  • Pollinator identification at flower stands
  • Pest/beneficial counterfactual checking (is this a parasitic wasp or a pest)
  • Tracking community observations of a specific weed across the season

Constraints

  • Rate limit — design caching and batching; don’t hammer the API per frame.
  • Computer Vision API accuracy — ~80% top-1, ~95% top-5 for common species; fails on rare or visually-similar species.
  • License — observation photos are licensed by uploader; respect the per-observation license before using for training.
  • Network dependency — needs internet; not for offline-field deployments without a Pi-side cache.

Source

See also

Auto-generated from this entry’s typed relations: frontmatter, grouped by relation type so the editorial signal isn’t flattened.

  • Member of: [[ingredient]]
  • Combines with: [[raspberry-pi]] · [[opencv]] · [[yolo]] · [[home-assistant]]

What links here, and how

Inbound connections from across the wiki, grouped by lens and by relationship. These appear automatically — every entity page declares what it links to, and that data populates here on the targets.

Practical

contains

1 inbound link · 5 outbound