Deep Learning to Accelerate Breeding for Crop Disease Resistance.
One of the aims of plant breeders is to breed plants with resistance to all significant diseases. In order to do this, much time is taken scoring the plants for the individual diseases.
Scoring can be a difficult task for several reasons including the similarity in appearance of many diseases, so requiring a highly trained plant pathologist for this job.
Megan is working on designing and training a deep learning network to help speed up and automate this process. Her model will be trained to identify and diagnose septoria and yellow rust in wheat plants from a photographed image of the plant.
The idea is to use this model in a mobile app, to be used in the field on a smartphone to speed up disease scoring.