CGIAR Wheat Growth Stage Challenge by CGIAR Platform for Big Data in Agriculture
The competition is hosted on Zindi. It had compeditors from around the world. Used it as a great learning opportunity. Its a real world business problem.
Picture-based insurance (PBI) improves crop insurance for small scale farmers around the world, where images from a smartphone camera keep a record of a crop’s growth and record any damage events that will affect insurance payouts. PBI is a great way for insurers to verify events and to monitor crop growth, but it can also generate overwhelming amounts of data once images stream in from thousands of farmers.
For this competition, I had to automate one part of the data processing pipeline: estimating the growth stage of a wheat crop based on an image sent in by a farmer. The images are automatically cropped to show a section of the field. My model must take in an image and output a prediction for the growth stage of the wheat shown, on a scale from 1 (crop just showing) to 7 (mature crop). My solution has to operate on the input image ONLY - no additional data may be used.
Reference points I used to build this model;
- FastAi Starter Notebook by @Johnowhitaker from Zindi
- Starter Notebook by @KarimAmer from Zindi
- Transfer learning for computer vision toturial from official PyTorch docs