News Arable
Is artificial intelligence conquering agriculture?
A PhD research project at the University of Groningen shows that great strides have been made in image classification in the field of deep learning, a form of artificial intelligence. This can come in handy in agricultural operations.
Deep learning is a technique based on multi-layered neural networks. In a sense, this form of artificial intelligence simulates the functioning of the human brain.
Recognizing, detecting and counting plants is of crucial importance in agricultural management. Recognizing, for example, diseases and weeds in a crop is necessary in order to be able to take the correct cultivation measures. Now this is mainly done through observation in the country. This is a time-consuming process that also requires a great deal of expertise from the farmer or crop advisor.
Computer vision and machine learning are now widely used techniques that can also be used for recognizing plants. Recognizing crops remains difficult with these techniques. The differences between, for example, a cultivated plant and a weed plant are small. In addition, the background and the quality of the images often change.
The technology does not stand still
On Tuesday 9 February, Pornntiwa Pawara will be awarded a doctorate at the University of Groningen for research into the various techniques that can be used for the recognition, detection and counting of plants. Her PhD research shows that recent developments in deep learning have made great strides in image classification.
In time, if the technique becomes reliable enough, these techniques can be applied, for example, to a field sprayer for site-specific weed control. But mechanical weed control or the mechanical selection of seed potatoes can also be considered.
Jurphaas Lugtenburg
© DCA Market Intelligence. This market information is subject to copyright. It is not permitted to reproduce, distribute, disseminate or make the content available to third parties for compensation, in any form, without the express written permission of DCA Market Intelligence.