TO PREDICT PLANT GROWTH AND YIELD IN GREEN ENVIRONMENT USING DEEP LEARNING

Abstract

Greenhouse growers and farmers in general rely heavily on accurate predictions of plant growth and harvest.
Improvements in environmental management, increased output, improved supply-and-demand matching, and
reduced costs may all result from the development of models that can properly simulate growth and yield.
Recent advances in ML, and especially Deep Learning (DL), may give robust new analytical tools. Researchers
seek to employ ML and DL methods to estimate production and plant growth variance in two greenhouse
settings: tomato yield forecasting and Ficus benjamina stem growth. In the prediction formulations, we use a
novel deep recurrent neural network (RNN) based on the Long Short-Term Memory (LSTM) neuron model. In
order to simulate the desired growth parameters, the RNN design takes into account both the historical values
for yield, growth, and stem diameter, as well as the microclimate circumstances. We propose a research that
uses the mean square error criteria to compare the results of several ML approaches, such as support vector
regression and random forest regression. Positive findings are given based on information collected from two
greenhouses in Belgium and the United Kingdom as part of the EU