Plant.id offers a plant identification service based on machine learning. Once you obtain the API key, you can use these client's code to speed-up the development of your implementation.

Send plant photos to our back end, wait for identification, and return the result. If the identification takes more than the identification_timeout, return identification info without any suggestions.

Send POST request to: https://api.plant.id/v2/identify and include following parameters:

• api_key- your API key
• images - one ore more images of the plant you want to identify (string - base64 or a file)

Other optional parameters:

• modifiers - list of strings:
  • "crops_simple"/"crops_fast" (default)/"crops_medium" - specify the speed & accuracy of the identification
  • "similar_images" - allow displaying of similar images -> If you want to get similar images in the response, you must include item similar_images here.
• plant_language - language code (ISO 639-1) used for plant_details (default "en")
• plant_details - list of strings, which determines which information about the plant will be included in the response (if the data is available)
  • "common_names" - list of common names of the plant in the language specified in plant_language
  • "url" - link to page with the plant profile (usually Wikipedia)
  • "name_authority" - scientific name of the plant
  • "wiki_description" - description of the plant from Wikipedia with source url and license
  • "taxonomy" - dictionary with the plant taxonomy
• and more (see the Documentation)

The result contains a list of suggestions of possible plant species (taxons). Each suggestion contains:

• scientific_name - the scientific name of the plant
• common_names - list of common names of the plant (if available)
• url - link to page with the plant profile (usually Wikipedia)
• wiki_description - description of the plant from Wikipedia (if available)
• taxonomy - taxonomy of the plant (if available)
• probability - certainty level that suggested plant is the one from the photo
• similar_images - representative images of the identified species carefully selected by the model, so it resembles the input image (Similar images are included in the result only if you add the value similar_image in the modifiers list of the request.)
• and more (see the Documentation)

We prepared a simple code to demostrate, how the API works. See the Python example

See our documentation for full reference.