Articles
Deep learning model for monitoring daily tomato plant growth
Article number
1337_38
Pages
283 – 288
Language
English
Abstract
The speaking plant approach (SPA), a sophisticated strategy for environmental control in greenhouses, has attracted a lot of attention.
The first and most important step in the SPA concept is to obtain physiological information from a plant and then to judge whether the plant is healthy.
The daily measurement of plant growth is important in the SPA for environmental control.
Various imaging techniques have been investigated for this task.
A robotized chlorophyll fluorescence (CF) imaging system that evaluates daily changes in the photosynthetic function of a tomato canopy was developed in our previous study and later commercialized.
In this study, for daily stem elongation measurement, we developed a deep learning model that detects a tomato plants shoot apex in a CF image.
The CF imaging system captures images by moving along the target tomato canopy at a constant speed.
Measurements were conducted for 65 consecutive days.
YOLOv3, a representative object detection algorithm, was used for the shoot apex detection model.
Shoot apexes captured in the centre of a CF image were manually annotated. 90% of the annotated data were used for training (including validation) the model and the remaining 10% of data were used for testing the model.
The developed deep learning model detected shoot apexes with an average F-measure of 0.99. This result indicates that the automatic measurement of daily stem elongation of tomato plants can be achieved by integrating the robotized CF imaging system and the developed deep learning model.
The first and most important step in the SPA concept is to obtain physiological information from a plant and then to judge whether the plant is healthy.
The daily measurement of plant growth is important in the SPA for environmental control.
Various imaging techniques have been investigated for this task.
A robotized chlorophyll fluorescence (CF) imaging system that evaluates daily changes in the photosynthetic function of a tomato canopy was developed in our previous study and later commercialized.
In this study, for daily stem elongation measurement, we developed a deep learning model that detects a tomato plants shoot apex in a CF image.
The CF imaging system captures images by moving along the target tomato canopy at a constant speed.
Measurements were conducted for 65 consecutive days.
YOLOv3, a representative object detection algorithm, was used for the shoot apex detection model.
Shoot apexes captured in the centre of a CF image were manually annotated. 90% of the annotated data were used for training (including validation) the model and the remaining 10% of data were used for testing the model.
The developed deep learning model detected shoot apexes with an average F-measure of 0.99. This result indicates that the automatic measurement of daily stem elongation of tomato plants can be achieved by integrating the robotized CF imaging system and the developed deep learning model.
Authors
S. Toda, T. Higuchi, T. Sakamoto, T. Kanoh, N. Fujiuchi, H. Nishina, K. Takayama
Keywords
chlorophyll fluorescence imaging, greenhouse, plant diagnosis, robot, speaking plant approach, stem elongation
Online Articles (57)