Using airborne video to map winter
weeds in emerging crops
Research Supervisor:
Dr David Lamb
Research Staff:
Mr Mark Weedon
Funding:
$4,000 (CRC WMS)
Project Summary:
Airborne multispectral imaging can be used to discriminate weeds against background crop
residue or soil. It is then possible to convert target images
into weed maps, once the image has been
geo-rectified using the location of known objects within the
image and the weeds have been suitable classified
using computer-based image processing. This project
aims to measure the potential accuracy of using
airborne multispectral imaging to generate weed maps with
the intention of using such maps to guide
on-ground precision weed spraying technology. The
accuracy will be quantified by comparing airborne-derived
weed maps directly with accurate on-ground weed
maps generated using a data-logger and differential
global positioning system (GPS). |