Tech Briefs Magazine - August 2024 - 7

Q & A
Increasing Agricultural Yields, Efficiency, and
Sustainability with AI
Professor Chris Reberg-Horton
and his
colleagues at North
Carolina State University's
Plant Sciences
Initiative are using
the university's Grace
Hopper 200 supercomputer
to create
AI for agricultural applications.
Tech Briefs: Could you describe your
project.
Chris Reberg-Horton: I'm an applied
agricultural scientist, so this project is applying
AI to my work. We've built a robot
named BenchBot 3.0, which takes hundreds
of thousands of images of all the
scenes you might see on a farmscape, all
the cash crops, weeds, and cover crops.
We've been imaging major weeds that
grow on farms in North America. We try
to image them every day from the second
they emerge until they get to be a decent
size. We also need more than one representative
of every species, because plants
are just like humans, everyone looks a little
bit different. We then create synthetic
training images, from images of plants
growing in pots and by also going into
the field and imaging plants there as well.
We extract from each of our images, the
image of a particular weed at a particular
stage of growth, for example, and label it.
We call that a virtual asset. We do the same
for different types of soil. We have hundreds
of thousands of these assets, which
we assemble in different combinations to
create synthetic scenes. And since each asset
has been labeled, we are able to identify
every pixel of the synthetic image. These
synthetic images are then used by our newly
acquired NVIDIA Grace Hopper supercomputer
to train the AI.
Tech Briefs: So, once AI has correctly
identified the synthetic image, what
do you do with that information?
Reberg-Horton: We want to build a massive
database that anyone, large or small,
Tech Briefs, August 2024
www.techbriefs.com
7
can use to train AI for their own specific
application. Here in America, and in
North Carolina in particular, weeds have
evolved resistance to herbicides . With
our database, we can help farmers gather
dynamic knowledge about how the weeds
are responding to their control tactics.
Tech Briefs: Does your application
also help with analyzing soil factors?
Reberg-Horton: It's critical to know
the nitrogen levels in the soil, but we
can't tell the grower exactly what nitrogen
rate a field needs - they have to use
their own map for that. We layer on top
of their map, a correction factor for the
nitrogen they have obtained by growing
legumes in their cover crops.
Reberg-Horton: The task for the AI is
making sure that any other plant like a
weed growing in there is separated from
the crop.
To read the full interview, visit www.
techbriefs.com.
Tech Briefs: How does the farmer
image his fields? What equipment is
used for it?
Reberg-Horton: We have built a camera
system that gets deployed on their
sprayers - we can put up to eight cameras
across the spray boom.
Tech Briefs: How does your
application help with cash crops?
https://www.techbriefs.com http://info.hotims.com/86258-705 https://www.techbriefs.com
Tech Briefs Magazine - August 2024

Table of Contents for the Digital Edition of Tech Briefs Magazine - August 2024
