Review: PlantNet Plant Identification App
PlantNet is a web and app-based citizen science platform that uses artificial intelligence (AI) to identify and inventory tens of thousands of plant species. Users can share observations which can then be reviewed by the PlantNet community and used by the underlying AI model to recognize plants. A sufficient degree of confidence is required in order for the observation to be added to the public database and the AI model. To take into account new observations and revisions, the AI model is regularly re-trained.
This diagram demonstrates the data flow of observations in the PlantNet system.
PlantNet has collected over 1 billion plant images and is one of the biggest contributors to the GBIF (Global Biodiversity Information Facility) international database. Data from GBIF can be downloaded and explored, or you can view the hundreds of publications that make use of PlantNet data. PlantNet's highly sophisticated process for identifying and validating observations has made them one of the most accurate plant identification applications.
I've been using PlantNet for several months to explore and learn more about the local plant species in the Sonoran Desert where I live. It's easy to use and I appreciate that it is not only a plant identification tool, but also a citizen science initiative producing useful data for research. One important note is that if you want to participate in the citizen science aspect of the platform, you must share your location as it is required for species mapping. If you'd like to learn more about using the app, this video provides a quick and straightforward tutorial.