Machine Learning and Computer Vision System for Phenotype Data Acquisition and Analysis in Plants.

Pedro J Navarro,Marcos Egea-Cortines,Julia Weiss,Fernando Pérez

doi:10.3390/s16050641

Pedro J Navarro, Marcos Egea-Cortines + Show 2 more

Open Access

https://doi.org/10.3390/s16050641

Copy DOI

Abstract

Phenomics is a technology-driven approach with promising future to obtain unbiased data of biological systems. Image acquisition is relatively simple. However data handling and analysis are not as developed compared to the sampling capacities. We present a system based on machine learning (ML) algorithms and computer vision intended to solve the automatic phenotype data analysis in plant material. We developed a growth-chamber able to accommodate species of various sizes. Night image acquisition requires near infrared lightning. For the ML process, we tested three different algorithms: k-nearest neighbour (kNN), Naive Bayes Classifier (NBC), and Support Vector Machine. Each ML algorithm was executed with different kernel functions and they were trained with raw data and two types of data normalisation. Different metrics were computed to determine the optimal configuration of the machine learning algorithms. We obtained a performance of 99.31% in kNN for RGB images and a 99.34% in SVM for NIR. Our results show that ML techniques can speed up phenomic data analysis. Furthermore, both RGB and NIR images can be segmented successfully but may require different ML algorithms for segmentation.

Highlights

The advent of the so-called omics technologies has been a major change in the way experiments are designed and has driven new ways to approach biology
We evaluated the results of training stage of the machine learning (ML) algorithms with a leave-one-out training sets obtained from RGB and near infrared (NIR) images
The third columnof shows after post-processing results after application the SVMresults classifier with quadratic function and with data normalisation dn1

Summary

Introduction

The advent of the so-called omics technologies has been a major change in the way experiments are designed and has driven new ways to approach biology. As a result biology has become a field where big data accumulates, and which requires analytical tools [1]. The latest newcomer in the field of automatic sampling is the so-called phenomics. It comprises any tool that will help acquire quantitative data of phenotypes. Plant growth and development can be considered as a combination of a default program that interacts with biotic and abiotic stresses, light and temperature to give external phenotypes. The outcome or end point, and kinetics and their changes is becoming increasingly important to understand plants as a whole and become more precise at experimental designs. One of the newest developments is automatic image acquisition [2]

Objectives

Results

Conclusion