UAV- and Random-Forest-AdaBoost (RFA)-Based Estimation of Rice Plant Traits

Derraz Radhwane,Muhamad Faiz Che Hashim,Siti Najja Mohd Zad,Farrah Melissa Muharam,Zed Zulkafli,Mohd Razi Ismail,Asniyani Nur Haidar Abdullah,Khairudin Nurulhuda,Mohamad Arif Tarmizi

doi:10.3390/agronomy11050915

Derraz Radhwane, Muhamad Faiz Che Hashim + Show 7 more

Open Access

https://doi.org/10.3390/agronomy11050915

Copy DOI

Journal: Agronomy	Publication Date: May 7, 2021
Citations: 19	License type: CC BY 4.0

Affiliation: Universiti Putra Malaysia

Abstract

Rapid, accurate and inexpensive methods are required to analyze plant traits throughout all crop growth stages for plant phenotyping. Few studies have comprehensively evaluated plant traits from multispectral cameras onboard UAV platforms. Additionally, machine learning algorithms tend to over- or underfit data and limited attention has been paid to optimizing their performance through an ensemble learning approach. This study aims to (1) comprehensively evaluate twelve rice plant traits estimated from aerial unmanned vehicle (UAV)-based multispectral images and (2) introduce Random Forest AdaBoost (RFA) algorithms as an optimization approach for estimating plant traits. The approach was tested based on a farmer’s field in Terengganu, Malaysia, for the off-season from February to June 2018, involving five rice cultivars and three nitrogen (N) rates. Four bands, thirteen indices and Random Forest-AdaBoost (RFA) regression models were evaluated against the twelve plant traits according to the growth stages. Among the plant traits, plant height, green leaf and storage organ biomass, and foliar nitrogen (N) content were estimated well, with a coefficient of determination (R2) above 0.80. In comparing the bands and indices, red, Normalized Difference Vegetation Index (NDVI), Ratio Vegetation Index (RVI), Red-Edge Wide Dynamic Range Vegetation Index (REWDRVI) and Red-Edge Soil Adjusted Vegetation Index (RESAVI) were remarkable in estimating all plant traits at tillering, booting and milking stages with R2 values ranging from 0.80–0.99 and root mean square error (RMSE) values ranging from 0.04–0.22. Milking was found to be the best growth stage to conduct estimations of plant traits. In summary, our findings demonstrate that an ensemble learning approach can improve the accuracy as well as reduce under/overfitting in plant phenotyping algorithms.

Highlights

In recent years, remote sensing platforms, especially aerial unmanned vehicle (UAV)and image processing methods, have been intensively explored as a preliminary step to plant phenotyping [1,2,3,4]
A consistent increasing pattern was observed throughout the season for the remaining of traits, i.e., plant height (PH), leaf color chart (LCC), dead leaf (DL), storage organ (SO) and TOTAL biomass
A gradual increasing trend was noticeable for PH and LCC (0.5 to 0.9 m and 3.0 to 3.6, respectively), while a much more rapid inclination was noticed for DL, SO and TOTAL biomass (2.6 to 26.3 g quadrant−1, 6.1 to 112.3 g quadrant−1 and 74.2 to 306.3 g quadrant−1, respectively)

Summary

Introduction

Remote sensing platforms, especially aerial unmanned vehicle (UAV)and image processing methods, have been intensively explored as a preliminary step to plant phenotyping [1,2,3,4]. Many recent studies [6,7,8,9,10,11,12,13,14,15,16,17] have demonstrated the capability of UAV based sensors to estimate common rice agronomics traits such as leaf or/and plant N content, the biomass of separate crop organs, yield components and other physiological responses that are useful for phenotyping studies. Fewer studies have been conducted for the rice leaf area index (LAI), plant height (PH), plant density and chlorophyll content [6,8,13,14,15,16,17]

Objectives

Methods

Results

Discussion

Conclusion