Leaf-Movement-Based Growth Prediction Model Using Optical Flow Analysis and Machine Learning in Plant Factory.

Shogo Nagano,Shogo Moriyuki,Hirokazu Fukuda,Hiroshi Mineno,Kazumasa Wakamori

doi:10.3389/fpls.2019.00227

Abstract

Productivity stabilization is a critical issue facing plant factories. As such, researchers have been investigating growth prediction with the overall goal of improving productivity. The projected area of a plant (PA) is usually used for growth prediction, by which the growth of a plant is estimated by observing the overall approximate movement of the plant. To overcome this problem, this study focused on the time-series movement of plant leaves, using optical flow (OF) analysis to acquire this information for a lettuce. OF analysis is an image processing method that extracts the difference between two consecutive frames caused by the movement of the subject. Experiments were carried out at a commercial large-scale plant factory. By using a microcomputer with a camera module placed above the lettuce seedlings, images of 338 seedlings were taken every 20 min over 9 days (from the 6th to the 15th day after sowing). Then, the features of the leaf movement were extracted from the image by calculating the normal-vector in the OF analysis, and these features were applied to machine learning to predict the fresh weight of the lettuce at harvest time (38 days after sowing). The growth prediction model using the features extracted from the OF analysis was found to perform well with a correlation ratio of 0.743. Furthermore, this study also considered a phenotyping system that was capable of automatically analyzing a plant image, which would allow this growth prediction model to be widely used in commercial plant factories.

Highlights

Closed-type plant factories, which cultivate plants in closed systems with controlled temperature, humidity, and light, are attracting attention as a new type of cultivation method, capable of producing the extra food needed to respond to population growth, while protecting the environment, improving health, and achieving economic growth (Kozai et al, 2015; Anpo et al, 2018; Kozai, 2018)
This study involved extraction of the image data features that are related to the leaf movement and the subsequent use of machine learning to construct a growth prediction model for lettuce, which is a typical crop grown in a closed-type plant factory
support vector regression (SVR) was performed to increase the accuracy of the prediction, for which there are no methods for explaining the importance of features such as gradient boost regression (GBR)

Summary

INTRODUCTION

Closed-type plant factories, which cultivate plants in closed systems with controlled temperature, humidity, and light, are attracting attention as a new type of cultivation method, capable of producing the extra food needed to respond to population growth, while protecting the environment, improving health, and achieving economic growth (Kozai et al, 2015; Anpo et al, 2018; Kozai, 2018). The authors’ group constructed a high-throughput growth prediction model for lettuce cultivars based on chlorophyll fluorescence for application to a commercial plant factory (Moriyuki and Fukuda, 2016). Machine learning is a promising technique for the analysis of large amounts of data and is mostly performed for prediction and classification tasks This method is widely used in various research fields, including plant production, plant science, and plant phenotyping (Moriyuki and Fukuda, 2016; Singh et al, 2016; Gutiérrez et al, 2018; Moghimi et al, 2018; Pineda et al, 2018; Zhang et al, 2018). This study involved extraction of the image data features that are related to the leaf movement and the subsequent use of machine learning to construct a growth prediction model for lettuce, which is a typical crop grown in a closed-type plant factory. The experiments were performed in an actual commercial large-scale plant factory with a daily output of 5,000 lettuces

MATERIALS AND METHODS

RESULTS

DISCUSSION