FORREST Research Foundation scholar Monica Danilevicz is on a mission to support crop breeders and growers make better-informed decisions in the field.
With guidance from her PhD supervisor professor Dave Edwards, who leads the Applied Bioinformatics Group at The University of Western Australia, Ms Danilevicz is leading two projects that use images collected by drones to train artificial intelligence models to accomplish their goals.
"The first project is aimed to help breeders select the highest-yielding varieties early in the field trials based on the plant phenotype a couple of months after sowing," Ms Danilevicz said.
"This tool has the potential to accelerate crop breeding, allowing researchers to direct resources to the most promising varieties and deliver better adapted crops faster."
The second project aims to develop a new weed detection model for Australian farmers.
"The focus is to locate and quantify weed density growing among similar looking crops, which will be used to guide weed management strategies that protect crop yield and reduce herbicide usage," Ms Danilevicz said.
With climate change affecting crop production and threatening food security worldwide, Ms Danilevicz hopes her research will help breeders develop climate-resilient crops with higher yields.
"I believe the second project will benefit canola and lupin growers to reduce yield loss, as it is estimated that weeds cost Australian grain growers $3.3 billion annually," she said.
"Herbicide researchers can also employ the weed detection model to quantify the effectiveness of the treatment in these species, which will be performed by one of our collaborators, AHRI."
Want weekly news highlights delivered to your inbox? Sign up to the Farm Weekly newsletter.