Machine Learning Methods for Automatic Segmentation of Images of Field and Glasshouse Based Plants for High Throughput Phenotyping

A - Papers appearing in refereed journals

Okyere, F., Cudjoe, D., Sadeghi-Tehran, P., Virlet, N., Riche, A. B., Castle, M., Greche, L., Mohareb, F., Simms, D. M., Mhada, M. and Hawkesford, M. J. 2023. Machine Learning Methods for Automatic Segmentation of Images of Field and Glasshouse Based Plants for High Throughput Phenotyping. Plants - Basel. 12 (10), p. 2035. https://doi.org/10.3390/plants12102035

AuthorsOkyere, F., Cudjoe, D., Sadeghi-Tehran, P., Virlet, N., Riche, A. B., Castle, M., Greche, L., Mohareb, F., Simms, D. M., Mhada, M. and Hawkesford, M. J.
Abstract

Image segmentation is a fundamental but critical step for achieving automated high- throughput phenotyping. While conventional segmentation methods perform well in homogenous environments, the performance decreases when used in more complex environments. This study aimed to develop a fast and robust neural-network-based segmentation tool to phenotype plants in both field and glasshouse environments in a high-throughput manner. Digital images of cowpea (from glasshouse) and wheat (from field) with different nutrient supplies across their full growth cycle were acquired. Image patches from 20 randomly selected images from the acquired dataset were transformed from their original RGB format to multiple color spaces. The pixels in the patches were annotated as foreground and background with a pixel having a feature vector of 24 color properties. A feature selection technique was applied to choose the sensitive features, which were used to train a multilayer perceptron network (MLP) and two other traditional machine learning models: support vector machines (SVMs) and random forest (RF). The performance of these models, together with two standard color-index segmentation techniques (excess green (ExG) and excess green–red (ExGR)), was compared. The proposed method outperformed the other methods in producing quality segmented images with over 98%-pixel classification accuracy. Regression models developed from the different segmentation methods to predict Soil Plant Analysis Development (SPAD) values of cowpea and wheat showed that images from the proposed MLP method produced models with high predictive power and accuracy comparably. This method will be an essential tool for the development of a data analysis pipeline for high-throughput plant phenotyping. The proposed technique is capable of learning from different environmental conditions, with a high level of robustness

KeywordsFeature extraction; Imaging; Machine learning; Phenotyping; Segmentation
Year of Publication2023
JournalPlants - Basel
Journal citation12 (10), p. 2035
Digital Object Identifier (DOI)https://doi.org/10.3390/plants12102035
Open accessPublished as ‘gold’ (paid) open access
FunderBiotechnology and Biological Sciences Research Council
Office Chérifien des Phosphate (OCP)
Funder project or codeDesigning Future Wheat - WP1 - Increased efficiency and sustainability
Image analysis for plant phenotyping - machine learning based methods for analysis of multi-model and multi-dimensional remote sensing data from high-throughput plant phenotyping
Publisher's version
Output statusPublished
Publication dates
Online19 May 2023
Publication process dates
Accepted10 May 2023
PublisherMDPI
ISSN2223-7747

Permalink - https://repository.rothamsted.ac.uk/item/98x20/machine-learning-methods-for-automatic-segmentation-of-images-of-field-and-glasshouse-based-plants-for-high-throughput-phenotyping

123 total views
33 total downloads
0 views this month
0 downloads this month
Download files as zip