Finding and retaining staff is more difficult than ever across a myriad of Australian workforces right now, including the agriculture sector.
The University of Southern Queensland has been taking on the challenge on behalf of the cotton industry to investigate how automated machine vision tools might be able to automate or augment some activities, such as crop scouting, to improve data collection and labour efficiency, boosting productivity and profitability.
The work will continue to identify target areas where machine vision tools can assist industry personnel.
Dr Alison McCarthy, a Senior Research Fellow (Irrigation and Cropping Systems) from the University’s Centre for Agricultural Engineering continues to lead this research.
“Elements of farming like scouting the field for pests and diseases or collecting plant measurements underpin everyday decisions for cotton farmers, from insect control decisions and crop rotations to irrigation management,” Dr McCarthy said.
“These are activities that are vital to successful everyday operations, but they are also heavily reliant on having enough people on the ground to carry them out.
“This research project will develop a novel suite of automated machine vision tools, from infield stationary cameras to smartphones, to make infield monitoring more efficient and to support agronomic activities.
“Crop agronomy is complex. These tools will provide agronomists and growers with consistent and accurate information.”
The research includes Dr Derek Long from University of Southern Queensland, as well as agronomists, entomologists and plant pathologists at Queensland's Department of Agriculture and Fisheries, with support from the Cotton Research and Development Corporation and Cotton Seed Distributors.
Dr McCarthy said exploring the feasibility of multiple machine vision systems is critical to getting a handle on how they can be utilised for a range of activities.
“We would hope to see outcomes such as improved repeatability of insect assessment using machine vision and more efficient collection of agronomic farm data to measure plant features that underpin decision making,” she said.
“We know too that detecting disease distribution is critical for planning crop management, so these tools should also offer data for better scheduling of crop rotation to curb disease.
"Data can also be used to support improved irrigation scheduling using thermal cameras for irrigation detection.
“Agronomists will be able to cover more hectares with a wider spread of sensors providing them timely site-specific information.”
