Monash University researchers have developed a new monitoring system that employs artificial intelligence (AI) to track bee movement to boost pollination and crop yield.
The study, published in the International Journal of Computer Vision, gathered data on over 2000 insect tracks at a commercial strawberry farm in Victoria by documenting pollinators such as honey bees, hover flies, moths, butterflies, and wasps.
The recordings were then examined using Computer Vision and AI to follow individual motions of individual insects, count them, and monitor their flower visits. This made it possible for farmers and researchers to comprehend various species’ role in pollination.
Monash University stated that the optimal number of pollinator visits to flowers is required for optimal pollination. Too few or too many visits, or visits by poor insect pollinators, can affect the quality of food produced by a flowering plant, hence reducing production.
According to research co-author NativeBee+Tech Facility Lab Director Associate Professor Alan Dorin, traditional insect monitoring techniques on farms are time-consuming, labour-intensive, and can yield inaccurate or unreliable data.
“The monitoring system developed through this study can generate same-day data of crop pollination levels and provide farmers the evidence they need to inform decision-making,” Associate Professor Dorin said.
Associate Professor Dorin stated that understanding how well a crop has been pollinated allows growers to change hive locations and numbers to increase pollination levels.
“Farmers might also open or close greenhouse sidewalls to encourage or discourage insect visits from particular directions. They may decide to add flowers to entice insects to explore crop regions that have not been pollinated adequately,” Associate Professor Dorin added.
He claimed that these basic actions could improve pollination rates and yields of market-quality fruit. He said they anticipate that this technology will serve as a model for future precision pollination research.
The researchers created customised software for the monitoring system to analyse the massive amount of data and reliably follow individual insects flying over dense foliage.
Dr Malika Ratnayake, the study’s principal researcher, stated that one of the most difficult challenges throughout the research was identifying the movement of different insects within a video so that the same insect path was not counted numerous times.
“The advanced software developed for the system combines AI-based object-detection capabilities with separate foreground detection algorithms to identify the precise positions of insects and the flowers they visit in the recorded videos,” Dr Ratnayake stated.
According to Dr Ratnayake, the software also incorporates capabilities that improve data processing efficiency and reduce computer power consumption.
“We have opted to keep this software open-source so it is accessible to anyone who wants to build similar monitoring systems or other applications to optimise and analyse different data points captured through videos,” Dr Ratnayake added.
In the future, the researchers plan to use the monitoring system to investigate the long-term effects and outcomes of precision pollination techniques and how they affect the quality of food production and output through multiple crop cycles.
The researchers will collaborate with the Australian Blueberry Growers Association, Costa Group’s berries division, CSIRO, Western Sydney University, and the University of New England to expand on this research.
Collaborations with the University of Trento in Italy and the German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig in Germany are also being explored.
The Australian Research Council (ARC) Discovery Projects grant, the Monash-Bosch AgTech Launchpad Primer Grant, AgriFutures, and the ARC Research Hub supported the research.
