Retrieval of Ryegrass Crude Protein from Spectral Data

Traditionally, these parameters are estimated through laboratory analysis. In this study a remote sensing approach was evaluated which employed a modified feature selection technique based on a genetic algorithm approach to determine the best spectral region, as well as the optimal and minimal numbers of bands when estimating crude protein. The statistical model was based on data collected from 266 full-range (350–2500 nm) proximal spectral measurements and corresponding ground truth observations in Australia and the Netherlands from May to November 2018.

The research project was executed as part of the PhD project of Gustavo Togeiro de Alckmin and was a cooperation between University of Tasmania, Tasmanian Institute of Agriculture-Centre for Dairy, Wageningen University, Wageningen Environmental Research and Wageningen Livestock Research.

The results of this research indicate that an affordable spectral-based sensor could estimate perennial ryegrass crude protein in outdoor environments using only eleven broad bands (10 nm bandwidth) within the visible to near-infrared range (400–1100 nm), provided that the protein is expressed in weight per area. Additionally, the models are transferable to new locations with a small decrease in performance (from 80 to 85 kg CP.ha^-1 RMSE). These results could lead to the development of sensors for autonomous deployment in an unmanned aerial or ground vehicle, providing end-users with essential information for the best agronomic practices.

The results of the study are published in the journal Remote Sensing in a paper entitled ‘Retrieval of Crude Protein in Perennial Ryegrass Using Spectral Data at the Canopy Level’

Abstract
Crude protein estimation is an important parameter for perennial ryegrass (Lolium perenne) management. This study aims to establish an effective and affordable approach for a non-destructive, near-real-time crude protein retrieval based solely on top-of-canopy reflectance. The study contrasts different spectral ranges while selecting a minimal number of bands and analyzing achievable accuracies for crude protein expressed as a dry matter fraction or on a weight-per-area basis. In addition, the model’s prediction performance in known and new locations is compared. This data collection comprised 266 full-range (350–2500 nm) proximal spectral measurements and corresponding ground truth observations in Australia and the Netherlands from May to November 2018. An exhaustive-search (based on a genetic algorithm) successfully selected band subsets within different regions and across the full spectral range, minimizing both the number of bands and an error metric. For field conditions, our results indicate that the best approach for crude protein estimation relies on the use of the visible to near-infrared range (400–1100 nm). Within this range, eleven sparse broad bands (of 10 nm bandwidth) provide performance better than or equivalent to those of previous studies that used a higher number of bands and narrower bandwidths. Additionally, when using top-of-canopy reflectance, our results demonstrate that the highest accuracy is achievable when estimating crude protein on its weight-per-area basis (RMSEP 80 kg.ha⁻¹). These models can be employed in new unseen locations, resulting in a minor decrease in accuracy (RMSEP 85.5 kg.ha⁻¹). Crude protein as a dry matter fraction presents a bottom-line accuracy (RMSEP) ranging from 2.5–3.0 percent dry matter in optimal models (requiring ten bands). However, these models display a low explanatory ability for the observed variability (R2 > 0.5), rendering them only suitable for qualitative grading.