menting the labor force.

“What we’re specializing in is be-
ing able to contact pieces of produce
and harvest them without damage,”
says Lessing. “The customers that I’m
working with, one of the major prob-
lems that they’re running into is they
can’t find enough labor to expand their
operations. So we’re supplementing
the labor force, and we’re delivering to
growing managers intelligence to every
piece of the operation.”
Another challenge faced by crop
growers is weed control. Tradition-
ally, farmers would spray herbicides
broadly across their crops, which not
only was wasteful, but also potentially
harmful to humans, as crops were
often overexposed to the chemicals.
Companies such as EcoRobotix, and
Blue River Technology with its See &
Spray agricultural machines equipped
with computer vision and machine
learning capabilities, claim they can
eliminate 90% of herbicide volumes
typically used on farms today.

The presence of weeds is not the only enemy of crops (and farmers). Plant diseases, if they are not detected quickly, can spread rapidly, and even incremental changes in the soil’s composition can have a drastic impact on crop yields.

“In terms of precision agriculture, [farms] are using more connected sensors on the ground to test nitrogen levels, for instance,” says Nisarg Desai, director of product management, IoT, GlobalSign, a networking technology company that has worked with agriculture companies to implement IoT communication security technology for plant sensor networks. Desai says IoT sensors are used to test soil moisture levels to identify flooding, overwatering, or ground freezing; Io T-enabled water and fertilizer delivery valves can also be remotely monitored and managed.

Some farms are turning to drone
technology, using unmanned aerial ve-
hicles (UAVs) equipped with a package
of high-definition cameras, IR sensors,
and image-recognition capabilities
to monitor crops, which can provide
significant increases in efficiency. In
a recent study, drone operator Pre-
cisionHawk found that farmers who
used drone-based aerial intelligence
instead of taking plot-based crop mea-
surements by hand were able to collect
data 2. 5 times more efficiently and 25%
more accurately, and the collection
itself was more objective, repeatable,
and standardized.

Thomas Haun, senior vice president of partnerships with South Carolina-based commercial drone and data company Precision Hawk, says drones provide a significant advantage over not only traditional ground-based visual inspections, but also over satellite-based inspection. From a drone flying overhead, sensors can monitor a variety of conditions, including plant yields and growth information, as well as identifying indications of disease or insect/animal damage, and even tracking temperatures.

“With the drone, you can go from vi-
sual data to multispectral data (image
data at specific frequencies), to ther-
mal data, to hyperspectral data (from
across the electromagnetic spectrum)
all in one flight,” Haun says, noting that
satellites are generally not equipped
with sensing technologies that allow a
very granular view of crops or plants.
“We’re capturing data at sub-centime-
ter resolution. There’s an actual spatial
resolution that our [sensors] are get-
ting, providing a real advantage.”
Currently, most drone operators
are limited by operational regulations,
which limit drone flights to those that
can be observed by the human operat-
ing the drone with his or her own eyes.
This generally limits drone operations
to about one square mile, according to
Haun, though companies can apply for
a Beyond Visual Line of Sight Waiver
from the U.S. Federal Aviation Admin-
istration (FAA), although few such
waivers are granted).

Precision Hawk has a waiver that allows the company to operate drones up to four miles away from an operator,

“With the drone, you
can go from visual
data to multispectral
data, to thermal data,
to hyperspectral data,
all in one flight.”

ACM
Member
News

AT THE INTERSECTION
OF CS AND
COMPUTATIONAL BIOLOGY
“I had an
analog
computer kit
when I was a
kid, where you
turned a dial
and it did
something,” says Dan Gusfield,
Distinguished Professor of
Computer Science at the
University of California, Davis
(UC Davis). Despite his
enchantment with that early
computer, Gusfield soon
learned his real attraction was
for discrete math, which offered
a natural segue into computer
science once he entered college.

Gusfield earned his undergraduate degree in computer science at the University of California at Berkeley, and his master’s degree in the same discipline from the University of California at Los Angeles. After receiving his Ph. D. in engineering science from the University of California at Berkeley, he spent six years as an assistant professor at Yale University, before moving to the University of California, Davis (UC Davis), where he has worked ever since.

He says his interests meet at the intersection of computer science and computational biology, an area on which he is writing his third book. Gusfield explains that the field of biology is becoming more quantitative, mathematical, and algorithmic, and these techniques are percolating down to biologists. In support of that, he is helping to establish an undergraduate major in quantitative biology at UC Davis.

With retirement on the horizon in a few years, Gusfield has no plans to end his career. He considers most of his academic work has been on computational techniques for problems that arise in biology, but he has never had the opportunity to take the next step and apply his techniques and programs to specific diseases.

“I want to delve more into a real disease,” he says.

—John Delaney