UAV-based imaging platform for monitoring maize growth throughout development

Field-based high-throughput phenotyping enhances phenomic and genomic predictions for grain yield and plant height across years in maize

Field-based phenomic prediction employs novel features, like vegetation indices (VIs) from drone images, to predict key agronomic traits in maize, despite challenges in matching biomarker measurement time points across years or environments. This study utilized functional principal component analysis (FPCA) to summarize the variation of temporal VIs, uniquely allowing the integration of this data into phenomic prediction models tested across multiple years (2018-2021) and environments. The models, which included 1 genomic, 2 phenomic, 2 multikernel, and 1 multitrait type, were evaluated in 4 prediction scenarios (CV2, CV1, CV0, and CV00), relevant for plant breeding programs, assessing both tested and untested genotypes in observed and unobserved environments. Two hybrid populations (415 and 220 hybrids) demonstrated the visible atmospherically resistant index's strong temporal correlation with grain yield (up to 0.59) and plant height. The first 2 FPCAs explained 59.3 ± 13.9% and 74.2 ± 9.0% of the temporal variation of temporal data of VIs, respectively, facilitating predictions where flight times varied. Phenomic data, particularly when combined with genomic data, often were comparable to or numerically exceeded the base genomic model in prediction accuracy, particularly for grain yield in untested hybrids, although no significant differences in these models' performance were consistently observed. Overall, this approach underscores the effectiveness of FPCA and combined models in enhancing the prediction of grain yield and plant height across environments and diverse agricultural settings.

Deciphering temporal growth patterns in maize: integrative modeling of phenotype dynamics and underlying genomic variations

Quantifying the temporal or longitudinal growth dynamics of crops in diverse environmental conditions is crucial for understanding plant development, requiring further modeling techniques. In this study, we analyzed the growth patterns of two different maize (Zea mays L.) populations using high-throughput phenotyping with a maize population consisting of 515 recombinant inbred lines (RILs) grown in Texas and a hybrid population containing 1090 hybrids grown in Missouri. Two models, Gaussian peak and functional principal component analysis (FPCA), were employed to study the Normalized Green-Red Difference Index (NGRDI) scores. The Gaussian peak model showed strong correlations (c. 0.94 for RILs and c. 0.97 for hybrids) between modeled and non-modeled temporal trajectories. Functional principal component analysis differentiated NGRDI trajectories in RILs under different conditions, capturing substantial variability (75%, 20%, and 5% for RILs; 88% and 12% for hybrids). By comparing these models with conventional BLUP values, common quantitative trait loci (QTLs) were identified, containing candidate genes of brd1, pin11, zcn8 and rap2. The harmony between these loci's additive effects and growing degree days, as well as the differentiation of RIL haplotypes across growth stages, underscores the significant interplay of these loci in driving plant development. These findings contribute to advancing understanding of plant-environment interactions and have implications for crop improvement strategies.

Multispectral-derived genotypic similarities from budget cameras allow grain yield prediction and genomic selection augmentation in single and multi-environment scenarios in spring wheat

With abundant available genomic data, genomic selection has become routine in many plant breeding programs. Multispectral data captured by UAVs showed potential for grain yield (GY) prediction in many plant species using machine learning; however, the possibilities of utilizing this data to augment genomic prediction models still need to be explored. We collected high-throughput phenotyping (HTP) multispectral data in a genotyped multi-environment large-scale field trial using two cost-effective cameras to fill this gap. We tested back to back the prediction ability of GY prediction models, including genomic (G matrix), multispectral-derived (M matrix), and environmental (E matrix) relationships using best linear unbiased predictor (BLUP) methodology in single and multi-environment scenarios. We discovered that M allows for GY prediction comparable to the G matrix and that models using both G and M matrices show superior accuracies and errors compared with G or M alone, both in single and multi-environment scenarios. We showed that the M matrix is not entirely environment-specific, and the genotypic relationships become more robust with more data capture sessions over the season. We discovered that the optimal time for data capture occurs during grain filling and that camera bands with the highest heritability are important for GY prediction using the M matrix. We showcased that GY prediction can be performed using only an RGB camera, and even a single data capture session can yield valuable data for GY prediction. This study contributes to a better understanding of multispectral data and its relationships. It provides a flexible framework for improving GS protocols without significant investments or software customization.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11032-024-01449-w.

UAV-based time-series phenotyping reveals the genetic basis of plant height in upland cotton

Plant height (PH) is an important agronomic trait affecting crop architecture, biomass, resistance to lodging and mechanical harvesting. Elucidating the genetic governance of plant height is crucial because of the global demand for high crop yields. However, during the rapid growth period of plants the PH changes a lot on a daily basis, which makes it difficult to accurately phenotype the trait by hand on a large scale. In this study, an unmanned aerial vehicle (UAV)-based remote-sensing phenotyping platform was applied to obtain time-series PHs of 320 upland cotton accessions in three different field trials. The results showed that the PHs obtained from UAV images were significantly correlated with ground-based manual measurements, for three trials (R2  = 0.96, 0.95 and 0.96). Two genetic loci on chromosomes A01 and A11 associated with PH were identified by genome-wide association studies (GWAS). GhUBP15 and GhCUL1 were identified to influence PH in further analysis. We obtained a time series of PH values for three field conditions based on remote sensing with UAV. The key genes identified in this study are of great value for the breeding of ideal plant architecture in cotton.

Row selection in remote sensing from four-row plots of maize and sorghum based on repeatability and predictive modeling

Remote sensing enables the rapid assessment of many traits that provide valuable information to plant breeders throughout the growing season to improve genetic gain. These traits are often extracted from remote sensing data on a row segment (rows within a plot) basis enabling the quantitative assessment of any row-wise subset of plants in a plot, rather than a few individual representative plants, as is commonly done in field-based phenotyping. Nevertheless, which rows to include in analysis is still a matter of debate. The objective of this experiment was to evaluate row selection and plot trimming in field trials conducted using four-row plots with remote sensing traits extracted from RGB (red-green-blue), LiDAR (light detection and ranging), and VNIR (visible near infrared) hyperspectral data. Uncrewed aerial vehicle flights were conducted throughout the growing seasons of 2018 to 2021 with data collected on three years of a sorghum experiment and two years of a maize experiment. Traits were extracted from each plot based on all four row segments (RS) (RS1234), inner rows (RS23), outer rows (RS14), and individual rows (RS1, RS2, RS3, and RS4). Plot end trimming of 40 cm was an additional factor tested. Repeatability and predictive modeling of end-season yield were used to evaluate performance of these methodologies. Plot trimming was never shown to result in significantly different outcomes from non-trimmed plots. Significant differences were often observed based on differences in row selection. Plots with more row segments were often favorable for increasing repeatability, and excluding outer rows improved predictive modeling. These results support long-standing principles of experimental design in agronomy and should be considered in breeding programs that incorporate remote sensing.

UAV Image-Based Crop Growth Analysis of 3D-Reconstructed Crop Canopies

Plant growth rate is an essential phenotypic parameter for quantifying potential crop productivity. Under field conditions, manual measurement of plant growth rate is less accurate in most cases. Image-based high-throughput platforms offer great potential for rapid, non-destructive, and objective estimation of plant growth parameters. The aim of this study was to assess the potential for quantifying plant growth rate using UAV-based (unoccupied aerial vehicle) imagery collected multiple times throughout the growing season. In this study, six diverse lines of lentils were grown in three replicates of 1 m² microplots with six biomass collection time-points throughout the growing season over five site-years. Aerial imagery was collected simultaneously with each manual measurement of the above-ground biomass time-point and was used to produce two-dimensional orthomosaics and three-dimensional point clouds. Non-linear logistic models were fit to multiple data collection points throughout the growing season. Overall, remotely detected vegetation area and crop volume were found to produce trends comparable to the accumulation of dry weight biomass throughout the growing season. The growth rate and G50 (days to 50% of maximum growth) parameters of the model effectively quantified lentil growth rate indicating significant potential for image-based tools to be used in plant breeding programs. Comparing image-based groundcover and vegetation volume estimates with manually measured above-ground biomass suggested strong correlations. Vegetation area measured from a UAV has utility in quantifying lentil biomass and is indicative of leaf area early in the growing season. For mid- to late-season biomass estimation, plot volume was determined to be a better estimator. Apart from traditional traits, the estimation and analysis of plant parameters not typically collected in traditional breeding programs are possible with image-based methods, and this can create new opportunities to improve breeding efficiency mainly by offering new phenotypes and affecting selection intensity.

Comparison of Remote Sensing Methods for Plant Heights in Agricultural Fields Using Unmanned Aerial Vehicle-Based Structure From Motion

Remote sensing using unmanned aerial vehicles (UAVs) and structure from motion (SfM) is useful for the sustainable and cost-effective management of agricultural fields. Ground control points (GCPs) are typically used for the high-precision monitoring of plant height (PH). Additionally, a secondary UAV flight is necessary when off-season images are processed to obtain the ground altitude (GA). In this study, four variables, namely, camera angles, real-time kinematic (RTK), GCPs, and methods for GA, were compared with the predictive performance of maize PH. Linear regression models for PH prediction were validated using training data from different targets on different flights ("different-targets-and-different-flight" cross-validation). PH prediction using UAV-SfM at a camera angle of -60° with RTK, GCPs, and GA obtained from an off-season flight scored a high coefficient of determination and a low mean absolute error (MAE) for validation data (R ² _val = 0.766, MAE = 0.039 m in the vegetative stage; R ² _val = 0.803, MAE = 0.063 m in the reproductive stage). The low-cost case (LC) method, conducted at a camera angle of -60° without RTK, GCPs, or an extra off-season flight, achieved comparable predictive performance (R ² _val = 0.794, MAE = 0.036 m in the vegetative stage; R ² _val = 0.749, MAE = 0.072 m in the reproductive stage), suggesting that this method can achieve low-cost and high-precision PH monitoring.

Individual Maize Location and Height Estimation in Field from UAV-Borne LiDAR and RGB Images

Crop height is an essential parameter used to monitor overall crop growth, forecast crop yield, and estimate crop biomass in precision agriculture. However, individual maize segmentation is the prerequisite for precision field monitoring, which is a challenging task because the maize stalks are usually occluded by leaves between adjacent plants, especially when they grow up. In this study, we proposed a novel method that combined seedling detection and clustering algorithms to segment individual maize plants from UAV-borne LiDAR and RGB images. As seedlings emerged, the images collected by an RGB camera mounted on a UAV platform were processed and used to generate a digital orthophoto map. Based on this orthophoto, the location of each maize seedling was identified by extra-green detection and morphological filtering. A seed point set was then generated and used as input for the clustering algorithm. The fuzzy C-means clustering algorithm was used to segment individual maize plants. We computed the difference between the maximum elevation value of the LiDAR point cloud and the average elevation value of the bare digital terrain model (DTM) at each corresponding area for individual plant height estimation. The results revealed that our height estimation approach test on two cultivars produced the accuracy with R2 greater than 0.95, with the mean square error (RMSE) of 4.55 cm, 3.04 cm, and 3.29 cm, as well as the mean absolute percentage error (MAPE) of 3.75%, 0.91%, and 0.98% at three different growth stages, respectively. Our approach, utilizing UAV-borne LiDAR and RGB cameras, demonstrated promising performance for estimating maize height and its field position.

Estimation of plant height and yield based on UAV imagery in faba bean (Vicia faba L.)

BACKGROUND

Faba bean is an important legume crop in the world. Plant height and yield are important traits for crop improvement. The traditional plant height and yield measurement are labor intensive and time consuming. Therefore, it is essential to estimate these two parameters rapidly and efficiently. The purpose of this study was to provide an alternative way to accurately identify and evaluate faba bean germplasm and breeding materials.

RESULTS

The results showed that 80% of the maximum plant height extracted from two-dimensional red-green-blue (2D-RGB) images had the best fitting degree with the ground measured values, with the coefficient of determination (R²), root-mean-square error (RMSE), and normalized root-mean-square error (NRMSE) were 0.9915, 1.4411 cm and 5.02%, respectively. In terms of yield estimation, support vector machines (SVM) showed the best performance (R² = 0.7238, RMSE = 823.54 kg ha^-1, NRMSE = 18.38%), followed by random forests (RF) and decision trees (DT).

CONCLUSION

The results of this study indicated that it is feasible to monitor the plant height of faba bean during the whole growth period based on UAV imagery. Furthermore, the machine learning algorithms can estimate the yield of faba bean reasonably with the multiple time points data of plant height.

Oat Crown Rust Disease Severity Estimated at Many Time Points Using Multispectral Aerial Photos

All plant breeding programs are dependent on plant phenotypic and genotypic data, but the development of phenotyping technology has been slow relative to that of genotyping. Crown rust (Puccinia coronata f. sp. avenae Erikss.) is the most important disease of cultivated oat (Avena sativa L.), making the development of disease-resistant oat cultivars an important breeding objective. Visual observation is the most common scoring method, but it can be laborious and subjective. We visually scored a diverse collection of 256 oat lines at a total of 27 time points in three disease nursery environments. Multispectral aerial photos were collected using an unmanned aerial vehicle at the same time points as the visual observations. The photos were analyzed, and subsets of the spectral properties of each plot were measured. Random forest modeling was used to model the relationship between the spectral properties of the plots and visually observed disease severity. The ability of the photo data and the random forest model to estimate visually observed disease severity was evaluated using three different cross-validation analyses. We specifically addressed the issue of assessing phenotyping accuracy across and within time points. The accuracy of the photo estimates was greatest for adult plants shortly before they began to senesce. Accuracy outside of that time frame was generally low but statistically significant. Unmanned aerial vehicle-mounted sensors could increase disease scoring efficiency, but additional investigation into the spectral signature of disease severity at all plant growth stages may be necessary to automate accurate full-season measurements.

Temporal Vegetation Indices and Plant Height from Remotely Sensed Imagery Can Predict Grain Yield and Flowering Time Breeding Value in Maize via Machine Learning Regression

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.

Canopy Roughness: A New Phenotypic Trait to Estimate Aboveground Biomass from Unmanned Aerial System

Cost-effective phenotyping methods are urgently needed to advance crop genetics in order to meet the food, fuel, and fiber demands of the coming decades. Concretely, characterizing plot level traits in fields is of particular interest. Recent developments in high-resolution imaging sensors for UAS (unmanned aerial systems) focused on collecting detailed phenotypic measurements are a potential solution. We introduce canopy roughness as a new plant plot-level trait. We tested its usability with soybean by optical data collected from UAS to estimate biomass. We validate canopy roughness on a panel of 108 soybean [Glycine max (L.) Merr.] recombinant inbred lines in a multienvironment trial during the R2 growth stage. A senseFly eBee UAS platform obtained aerial images with a senseFly S.O.D.A. compact digital camera. Using a structure from motion (SfM) technique, we reconstructed 3D point clouds of the soybean experiment. A novel pipeline for feature extraction was developed to compute canopy roughness from point clouds. We used regression analysis to correlate canopy roughness with field-measured aboveground biomass (AGB) with a leave-one-out cross-validation. Overall, our models achieved a coefficient of determination (R ²) greater than 0.5 in all trials. Moreover, we found that canopy roughness has the ability to discern AGB variations among different genotypes. Our test trials demonstrate the potential of canopy roughness as a reliable trait for high-throughput phenotyping to estimate AGB. As such, canopy roughness provides practical information to breeders in order to select phenotypes on the basis of UAS data.