Our technology

To provide the best results of our plant identification APIour tech stack uses Google’s Tensorflow and a python-based heavy lifting. Minor part of the service is based on Google Cloud and Microsoft Azure.

The neural network’s architecture is based on the Inception model, but we have implemented a bunch of customizations, for example to estimate identification certainty, select best adaptive representative photos and improve the accuracy with meta attributes like GPS coordinates.

Model specifications

0
Identifiable plant species
0 %
Success rate

Our newest model can identify 10 997 plant species. The set covers the most commonly requested plants from the United States, European Union, India, and Australia.

We are continually testing model accuracy. The average success rate ranges between 80%–90%, although it strongly depends on photo and plant quality. Our goal is to improve the average over 90% while extending the spectrum of identified species.

Roadmap

  • Object detection to be able to provide bouding boxes for different plant species within a single image
  • Photo quality estimation to enable the client to decide if the photo is suitable before it is submitted for identification; the most straightforward question would be: does the photo actually displays a plant?
  • Feature extraction