Skoltech Scientists Use Computer Vision and Machine Learning to Predict Plant Growth

Skolkovo Institute of Science and Technology researchers havedeveloped a new way to predict how a plant will grow, based on two-dimensional and three-dimensional images. Researchers at the Skolkovo Institute of Science and Technology (Skoltech) in Russia have developed a method for predicting increasing biomass of a plant using two-dimensional (2D) and three-dimensional (3D) images, which they say will improve the efficiency of precision farming. The research team initially measured plant growth in an artificial soil-free system with a 3D camera, and established the relationship between growth in total biomass and the expansion of the total surface area of a plant's leaves. They then measured the increase in total leaf area with a 2D camera and built a dynamic plant growth model based on the data. Using both 2D and 3D cameras eliminates time-consuming multi-parameter calculations.