University of Zurich deploys AI to observe plants

A team from the University of Zurich (UZH), supported by the university’s research focus on evolution, has been observing the behavior of plants in response to changed environmental conditions in a project. For this, a vast volume of data was evaluated using software supported by Artificial Intelligence (AI), further details of which can be found in a press release. As part of the PlantServation project, the researchers observed plants in a test field located at the university’s Irchel campus, with high-resolution image capture hardware used in tandem with AI-powered software. The plants were observed over three growing seasons (five months from the spring to fall), during which huge volumes of data were evaluated. Conventional methods are ill-suited to analysis of this kind, the press release explains. In addition, the methods conventionally used up to this point involve taking plant samples and destroying plant parts, which can in turn distort the test results.

Plants of the Arabidopsis genus were the focus of the researchers’ observations. In total, four million bird’s eye images of these plants were processed by way of machine learning technology. In this context, the researchers sought to record data in relation to the species-specific accumulation of reddish plant pigments known as anthocyanins in response to seasonal and annual fluctuations in temperature, light and precipitation.

PlantServation was developed in the experimental garden at UZH. “It was crucial for us to be able to use the garden on Irchel Campus to develop PlantServation’s hardware and software”, comments Kentaro Shimizu, lead author and co-director of the research focus, in the press release. However, its application goes further still, in that when it is combined with solar power, the hardware can be used even in remote sites, he explains. “With its economical and robust hardware and open-source software, PlantServation paves the way for many more future biodiversity studies that use AI to investigate plants other than Arabidopsis – from crops such as wheat to wild plants that play a key role for the environment”, Shimizu concludes. ce/ww