Abstract
Unoccupied aerial system (UAS; i.e., drone equipped with sensors) field-based high-throughput phenotyping (HTP) platforms are used to collect high quality images of plant nurseries to screen genetic materials (e.g., hybrids and inbreds) throughout plant growth at relatively low cost. In this study, a set of 100 advanced breeding maize (Zea mays L.) hybrids were planted at optimal (OHOT trial) and delayed planting dates (DHOT trial). Twelve UAS surveys were conducted over the trials throughout the growing season. Fifteen vegetative indices (VIs) and the 99th percentile canopy height measurement (CHMs) were extracted from processed UAS imagery (orthomosaics and point clouds) which were used to predict plot-level grain yield, days to anthesis (DTA), and silking (DTS). A novel statistical approach utilizing a nested design was fit to predict temporal best linear unbiased predictors (TBLUP) for the combined temporal UAS data. Our results demonstrated machine learning-based regressions (ridge, lasso, and elastic net) had from 4- to 9-fold increases in the prediction accuracies and from 13- to 73-fold reductions in root mean squared error (RMSE) compared to classical linear regression in prediction of grain yield or flowering time. Ridge regression performed best in predicting grain yield (prediction accuracy = ~0.6), while lasso and elastic net regressions performed best in predicting DTA and DTS (prediction accuracy = ~0.8) consistently in both trials. We demonstrated that predictor variable importance descended towards the terminal stages of growth, signifying the importance of phenotype collection beyond classical terminal growth stages. This study is among the first to demonstrate an ability to predict yield in elite hybrid maize breeding trials using temporal UAS image-based phenotypes and supports the potential benefit of phenomic selection approaches in estimating breeding values before harvest.
Highlights
This study aimed to (i) predict the temporal breeding values of maize hybrids using fifteen vegetative indices (VIs) and canopy height measurement (CHMs) derived from twelve Unoccupied aerial systems (UAS) flights over the two trials; (ii) compare performances of linear regression and machine learning based regressions in predicting grain yield and flowering times from temporal breeding values of fifteen VIs and CHMs; and (iii) calculate the temporal variable importance scores of the predictors in the best performing
This study revealed that flowering times (DTA and days to silking (DTS)) can be best predicted before flowering using the temporal VIs, which is critical where earliness traits for maize can be selected based on the values of certain VIs at specific early time points, such as normalized green red difference index (NGRDI)
Our findings suggested that temporal variations of VIs and CHMs should be trained together in a predictive model using machine learning based regressions to predict grain yield with more accuracy, when the number of predictors are high, which otherwise may lead to overfitting and multicollinearity problems
Summary
Estimating breeding values for genotypes (pedigrees) before harvest would be useful for speeding the breeding cycle while reducing combine measurement resources. Season breeding values can be enabled by implementing temporal field-based high throughput phenotyping (HTP) [1]. Unoccupied aerial systems (UAS; i.e., drones with sensor payloads) can objectively capture the temporal variation of complex traits in crops with limited resources; providing a new tool to dissect complex traits in a time-series manner. UAS provides higher quality raw images with better resolutions than traditional remote sensing platforms such as Landsat-derived images [2]. UAS raw images are processed (e.g., orthomosaic and point cloud densification) to generate geographically corrected mosaics over
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call