ADVANCES in remote sensing and data collection technologies that enable every individual plant in a paddock to be identified and phenotyped are set to open a new world of opportunity for researchers and agronomists.

Mitch Tuinstra

Engineers at Purdue University in Indiana, United States, are part-way through a project aimed at developing an automated, multiple-sensor system to locate, count and phenotype individual plants.

Speaking at the TropAg 2017 conference in Brisbane, Purdue University professor of plant breeding and genetics, Mitch Tuinstra, said the project’s brief was to come up with a high-throughput system for collecting and processing phenotyping data that would make data collection much easier, more effective and much less labour intensive.

“The major focus of our research has been developing technical solutions that allow us to create very precise remote sensing data of plants growing in field environments,” he said.

“There are tools that are currently publicly available that are widely used in terms of processing data, but many of them have serious problems with regards to the quality of the data. So, the group of engineers I have been working with have been trying to resolve some of those issues to get better quality data.”

Dr Tuinstra said much of the focus of the development work had been on the system’s application in crop trial work.

“One of the major areas it is leading us is the ability to collect extremely well resolved and geo-referenced data sets using different sensors and collected at different time points,” he said.

“It is going to open a window in terms of allowing us to look at plant development over time in ways that haven’t been possible before, and potentially to develop better predictive models that have better predictive capacity in terms of predicting productivity.”

Dr Tuinstra said the capacity to identify every plant in a field opened up new opportunities throughout the plant breeding and cropping industries.

“Our agronomy farm at Purdue University is 1400 acres. Our goal when we started this was that within 10 years we wanted to be able to identify and phenotype every plant in every field on the farm and align that data with any genetic information that might exist with those plants,” he said.

“That has tremendous potential for plant breeding and crop development, but also for the agricultural researchers. Often there is a huge amount of labour and effort required to collect data in agronomic trials. “

Dr Tuinstra said the new technology would not only have application in the research field, but also throughout the broadacre farming environment.

“Early on it will be picked up by the private sector such as crop management companies that can use the tools to make their lives more efficient,” he said.

“In the slightly longer term, large farmers are going to be using it as part of their integrated management practices using drone technologies to patrol their farms.”

Grain Central: Get our free daily cropping news straight to your inbox – Click here