Proceedings

Find matching any: Reset

Add filter to result:

Development of Variable Rate Applicators Using Real-Time Machine Vision Sensing and Control System for Spot-Application of Agrochemicals

Q. Zaman

NSAC Canada

The variable rate applicators comprised of a real-time sensing and control system were developed and tested for spot-application of agrochemicals (fertilizer and pesticides). Growers apply herbicides to control competing weeds, fungicides to control diseases and inorganic fertilizer to encourage production. The long-term objectives of this project were to develop innovative technologies which do not use prescription maps, but rely on sensors/digital cameras to provide real-time weed or plant detection information for agrochemical application in the specific section of the boom where the target (weeds, bare spots or plants) have been detected. The machine vision and control system of VR applicators consisted of μEye digital color cameras mounted on a separate boom in front of the tractor, fast computer, computerized VR controllers, Mid-Tech Legacy 6000 controller, flow valve, and solenoid valves were interfaced to a Pocket PC using wireless Bluetooth^® radio with compatible software. Real-time image processing software was developed in C⁺⁺ using Visual Studio 2010 to discriminate weeds and bare spots from plants. The digital color images taken by the cameras were processed in real-time. The final result of target detection in each image, converted to 5 VDC (ON/OFF) signal, was sent to VRC through U3-HV i/o unit to open the valve and nozzle for agrochemical application in the specific section of the boom where the target had been detected. The performance accuracy of real-time machine vision sensing and control system was evaluated for real-time detection and agrochemical application on right targets to reduce agrochemical usage and environmental pollution.

Keyword: controllers, digital color camera, machine vision, VR spreader, VR Sprayer, wild blueberry

Download paper