Cassava as a world food security crop still suffers from an inadequate means to measure early storage root bulking (ESRB), a trait that describes early maturity and a key characteristic of improved cassava varieties. The objective of this study is to evaluate the capability of ground penetrating radar (GPR) for non-destructive assessment of cassava root biomass. GPR was evaluated for this purpose in a field trial conducted in Ibadan, Nigeria. Different methods of processing the GPR radargram were tested, which included time slicing the radargram below the antenna surface in order to reduce ground clutter; to remove coherent sub-horizontal reflected energy; and having the diffracted energy tail collapsed into representative point of origin. GPR features were then extracted using Discrete Fourier Transformation (DFT), and Bayesian Ridge Regression (BRR) models were developed considering one, two and three-way interactions. Prediction accuracies based on Pearson correlation coefficient (r) and coefficient of determination (R²) were estimated by the linear regression of the predicted and observed root biomass. A simple model without interaction produced the best prediction accuracy of r = 0.64 and R² = 0.41. Our results demonstrate that root biomass can be predicted using GPR and it is expected that the technology will be adopted by cassava breeding programs for selecting early stage root bulking during the crop growth season as a novel method to dramatically increase crop yield.

This study explores the efficacy of utilizing a novel ground penetrating radar (GPR) acquisition platform and data analysis methods to quantify peanut yield for breeding selection, agronomic research, and producer management and harvest applications. Sixty plots comprising different peanut market types were scanned with a multichannel, air-launched GPR antenna. Image thresholding analysis was performed on 3D GPR data from four of the channels to extract features that were correlated to peanut yield with the objective of developing a noninvasive high-throughput peanut phenotyping and yield-monitoring methodology. Plot-level GPR data were summarized using mean, standard deviation, sum, and the number of nonzero values (counts) below or above different percentile threshold values. Best results were obtained for data below the percentile threshold for mean, standard deviation and sum. Data both below and above the percentile threshold generated good correlations for count. Correlating individual GPR features to yield generated correlations of up to 39% explained variability, while combining GPR features in multiple linear regression models generated up to 51% explained variability. The correlations increased when regression models were developed separately for each peanut type. This research demonstrates that a systematic search of thresholding range, analysis window size, and data summary statistics is necessary for successful application of this type of analysis. The results also establish that thresholding analysis of GPR data is an appropriate methodology for noninvasive assessment of peanut yield, which could be further developed for high-throughput phenotyping and yield-monitoring, adding a new sensor and new capabilities to the growing set of digital agriculture technologies.

Challenges in rapid prototyping are a major bottleneck for plant breeders trying to develop the needed cultivars to feed a growing world population. Remote sensing techniques, particularly LiDAR, have proven useful in the quick phenotyping of many characteristics across a number of popular crops. However, these techniques have not been demonstrated with cassava, a crop of global importance as both a source of starch as well as animal fodder. In this study, we demonstrate the applicability of using terrestrial LiDAR for the determination of cassava biomass through binned height estimations, total aboveground biomass and total leaf biomass. We also tested using single LiDAR scans versus multiple registered scans for estimation, all within a field setting. Our results show that while the binned height does not appear to be an effective method of aboveground phenotyping, terrestrial laser scanners can be a reliable tool in acquiring surface biomass data in cassava. Additionally, we found that using single scans versus multiple scans provides similarly accurate correlations in most cases, which will allow for the 3D phenotyping method to be conducted even more rapidly than expected.