Can Machine Learning Help In The Battle Against Ever-Tougher Weeds?

MARY HIGHTOWER

FAYETTEVILLE, ARKANSAS

   As weeds become more adept at evading control, researchers in Arkansas and elsewhere are testing whether machine learning can improve weed management and save farmers money.

   Jason Norsworthy, distinguished professor of weed science with the University of Arkansas System Division of Agriculture, has been evaluating the combination of image recognition software and machine learning that comprises John Deere’s See & SprayTM Ultimate system.

   Norsworthy will discuss his research and a demonstration of the See & SprayTM Ultimate will be provided during a field day set for Aug. 23 at the Northeast Research and Extension Center at Keiser in northeast Arkansas.

   “I’m in the second year of testing the machine learning in this system,” Norsworthy said. “This field day will give our Mid-South growers an opportunity to see it in action compared to a standard broadcast system.”

   The field day opens at 9:30 a.m., with field tours going until 11:15 a.m. The action will head indoors for lunch and with a panel discussion and Q&A with Norsworthy; William Patzoldt, director of agronomy for Blue River Technology, which led the software development; a farmer familiar with the system; and a representative from John Deere.

   The field tours include a demonstration of the technology, tips for herbicide savings while maintaining effective control; effect of sensitivity settings on detection and spray application and the effectiveness of the See & SprayTM Ultimate weed control in cotton and soybean production.

   Pre-event registration incentive

   Attendees are asked to register in advance. Pre-event registrants will be eligible to be one of 10 people selected to ride in the cab of the See & SprayTM Ultimate sprayer following lunch. The registration deadline is Aug. 19. Registration is available online. ∆

   MARY HIGHTOWER: University of Arkansas

 Norsworthy’s work is part of the Arkansas Agricultural Experiment Station, the research side of the University of Arkansas System Division of Agriculture.

 The See & Spray ™ system at work Image courtesy John Deere.