Ground-based visual assessments of co-occurring foliar diseases are time-consuming, laborious, and subjective due to the spatiotemporal overlapping of different lesion types and patterns. We took advantage of this scenario to explore the feasibility of unmanned aircraft systems (UAS)-derived multispectral vegetation indices to measure the variable incidence and severity of a mix of diseases. We rated separately the disease severity (as percent DLA or AUDPC) of artificially inoculated northern leaf blight (NLBart) along with naturally occurring northern leaf spot (NLSnat) and anthracnose leaf blight (ALBnat) in near-isogenic inbred (NILinbreds) and single-cross hybrid (NILhybrids) lines in Aurora, NY in 2018 ad 2019. NLBart and ALBnat were also scored in a contiguous field with a population of maize hybrids with broad genetic base. Total disease severity (tDSground) was estimated from the sum of the scored diseases. Disease severity and grain yield (GYground) were recorded from replicated 2-row plots. Two or three asynchronous UAS flights (no overlapping with ground-based visual estimates of each disease severity) were conducted in each crop season and plot-level vegetation indices (VIsair) were extracted from UAS-derived orthomosaics. Goodness of fit (R^2) between VIsair and tDSground were low (0-0.3) in the three germplasm groups. R^2 values between GYground and VIsair were higher (0.2-0.8) than those between GYground and tDSground (0.1-0.4). Our preliminary results highlight the challenges of dealing with a realistic field situation where the uncertain dynamics of a mix of pathogens and the contrasting perspectives (air vs. ground) involved in the disease screening add complexity that needs to be studied.

The traditional method of measuring the lettuce height is a manual measurement with instruments, which is greatly affected by human error.At present, researchers have proposed to use color cameras to obtain RGB images of lettuce, and to obtain the height of lettuce from the images. However, these tasks usually require camera calibration or a reference object with a known height, which is somewhat restrictive. Considering that deep neural networks have a powerful ability to feature extraction and expression, without camera calibration and reference objects, we try to use four networks of image recognition to explore the effect of deep learning on abstracting the lettuce height from RGB images. On the test set, including 80 images and height from 0.9 cm to 7.5 cm, we achieve a good result with a mean absolute error of 1.22 mm.

In order to count soybean seeds quickly and accurately, improve the speed of seed test and the level of soybean breeding, a method of soybean seed counting based on VGG-Two (VGG-T) was developed in this research. Firstly, in view of the lack of available image dataset in the field of soybean seed counting, a fast target point labeling method of combining pre-annotation based on digital image processing technology with manual correction annotation was proposed to speed up the establishment of publicly available soybean seed image dataset with annotation. Only 197 min were taken to mark 37,563 seeds when using this method, which saved 1592 min than ordinary manual marking and could reduce 96% of manual workload. At the same time, the dataset in this research is the largest annotated data set for soybean seed counting so far. Secondly, a method that combined the density estimation-based and the convolution neural network (CNN) was developed to accurately estimate the seed count from an individual threshed seed image with a single perspective. Thereinto, a CNN architecture consisting of two columns of the same network structure was used to learn the mapping from the original pixel to the density map. Due to the very limited number of training samples and the effect of vanishing gradients on deep neural networks, it is not easy for the network to learn all parameters at the same time. Inspired by the success of pre-training, this research pre-trained the CNN in each column by directly mapping the output of the fourth convolutional layer to the density map. Then these pre-trained CNNs were used to initialize CNNs in these two columns and fine-tune all parameters. Finally, the model was tested, and the effectiveness of the algorithm through three comparative experiments (with and without data enhancement, VGG16 and VGG-T, multiple sets of test set) was verified, which respectively provided 0.6 and 0.2 mean absolute error (MAE) in the original image and patch cases, while mean squared error (MSE) were 0.6 and 0.3. Compared with traditional image morphology operations, ResNet18, ResNet18-T and VGG16, the method proposed improving the accuracy of soybean seed counting. In the testset containing soybean seeds of different densities, the error fluctuation was small, and it still had excellent counting performance. At the same time, compared with manual counting and photoelectric seed counter, it saved about 2.493 h and 0.203 h respectively for counting 11,350 soybean seeds, realizing rapid soybean seeds counting.

Sorghum is an important cereal crop grown across the globe for its grain and biomass value. It can also efficiently use resources such as nitrogen, and multiple varieties that are nitrogen-use and light-capture efficient are constantly being developed. This study focuses on using the spectral signature of sorghum varieties to predict flowering days, which could be used as a proxy for plants’ growth/productivity and development trends, thus helping breeders make quick decisions about what varieties to move to the next stage. Multiple sorghum varieties from the sorghum association panel were planted in a replicate-design field experiment with the variable supply of nitrogen. The flowering days were monitored and recorded. The hyperspectral reflectance data were collected and used to build a sorghum flowering days predictive model. Although regression models such as partial least square have been used to predict plants’ phenotypes, the non-parametric ensemble machine learning model turned out to perform better on flowering days with an accurate model up to 5 days.

Maize (Zea mays L.) is one of the most consumed grains in the world and improving maize yield is of great importance for food security, especially under global climate change and more frequent severe droughts. However, traditional phenotyping methods relying on manual assessment are time-consuming and prone to human errors. Recently, the application of unmanned aerial vehicles (UAVs) has gained increasing attention in plant phenotyping due to their efficiency in data collection. Moreover, hyperspectral sensors integrated with UAVs can offer data streams with high spectral and spatial resolutions, which are essential for estimating plants’ physiological and biochemical traits. In this study, we developed machine learning models to estimate grain yield and flowering time of maize breeding lines using multi-temporal UAV-based hyperspectral imagery. The performance of multiple machine learning models and the efficacy of different hyperspectral features were evaluated on Genomes to Fields (G2F) experimental sites in Wisconsin. Results showed that ridge regression is the most robust model in estimating grain yield and flowering time, compared to random forest and support vector regression models. Furthermore, the ridge regression model achieved a correlation coefficient (r) of 0.551 for yield, 0.906 for days to silking, and 0.914 for days to anthesis when using the full-bands spectra features for estimation. In addition, we assessed the modeling performance using data acquired from different growing stages. The best time of applying the UAV survey was also identified in order to reduce the data collection efforts.

Sanbon Chaka Gosa1, Bogale Abebe Gebeyo12, Ravitejas Patil1, Ramón Mencia1, Dani Zamir1, Menachem Moshelion1# 1 The R.H. Smith Institute of Plant Sciences and Genetics in Agriculture, The R.H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, Rehovot, 76100 Israel 2 Current address: Department of Horticulture, College of Agriculture and Natural Resource, Dilla University, Dilla, Ethiopia #Corresponding Author Abstract Plant productivity in general and under stress, in particular, is a complex and comlitative trait largely influenced by environmental conditions. Among the most important traits are the whole-plant water-balance regulation mechanisms that dynamically change in order to maximize the metabolic activity of the plant. Due to the difficulty of high-throughput phenotyping of these physiological traits (e.g. transpiration, stomatal conductance, and photosynthesis), they are usually measured in static conditions or modeled based on only a few measuring points (low resolution). To overcome this challenge, we utilized a high-throughput gravimetric functional-phenotyping platform (PlantArray) along with a practical reverse phenotyping approach. We selected 30 tomato lines from multiple years of field yield data and functionally phenotyped them for their dynamic response curves using a variety of stress scenarios implemented using drought conditions (each plant received irrigation based on the amount of water it transpired). Our results show that resilient and tolerable traits, in the field, are associated with stomatal plastic conductance, i.e., maximum under well-irrigation, yet the rapid response to changes in environmental conditions (soil and atmospheric). The plastic traits of the idiotype lines were shown to increase water use efficiency (momentarily), thus maximizing yield in water deficit conditions. Based on manual characterizations of the idiotypes, it has been found that their abaxial surfaces have a greater density of stomata and a higher aperture during the early morning. Additionally, these lines showed rapid recovery after a drought. Our study concluded that reverse functional phenotyping can significantly reduce the pre-breading processes for yield-related traits. Keywords: Functional phenotyping, crops yield, —dynamic response, drought stress, stomatal conductance, reverse phenomics

Traditionally, walnut kernel color is scored by a human on a four-point color scale ranging from extra light to Amber, developed by the California DFA. Two important reasons for using a high throughput machine vision system for kernel phenotyping is to obtain better quantitative resolution to use in molecular breeding and to avoid errors in human vision phenotyping. RGB and LAB values give much more depth to qualitative calculations, and machine data is far more consistent than human scoring. Previous studies on walnut kernel imaging utilized a thresholding-based approach to segment walnut kernels from their background, however, detection accuracy can be improved. In this study, we make changes to detection methods, primarily using PyTorch computer vision processing and an improved thresholding method to better segment walnut kernels. The PyTorch CNN pipeline allows us to use specific photos to train a model, and the model can segment photos without the need to figure out image thresholds. Our proposed thresholding method uses the magick package in R instead of the ImageJ macros that were previously used. After taking photos with a CVS, we used the CNN model as well as an R script to identify and segment kernels from the background. Our preliminary data shows that with enough training, the CNN model is more robust in edge cases where we have overlapping kernels or shifted images. This paper will focus on exploring the differences between the three thresholding methods, and use the best method for future breeding projects.

Continuous collection and analysis of high-resolution phenotype data is critical to develop crops resilient to the consequences of climate change. Though web-accessible tools for parallel, reproducible scientiSic workSlows render big data increasingly tractable, software for plant science remains inadequate for large-scale precision agriculture. Cyberinfrastructure must present minimal barriers to entry, accommodate rapidly changing dependencies, support a wide variety of use cases, and weave together sensors at the edge, laptops, clusters, and cloud storage into a coherent virtual workspace. PlantIT is a web portal intended as such an environment. Platforms like PlantIT and its precursor DIRT [1] permit efSicient phenotyping and equip geographically distributed researchers with a code-optional interface. WorkSlows are published in Docker images, deployed as Singularity containers to public or private computing resources, and monitored in real time. Data are stored automatically in the CyVerse Data Store and can be annotated according to the MIAPPE [2] standard. GitHub integration provides versioning and repositories can be activated with a single conSiguration Sile, like Travis or GitHub Actions. Containers allow for a range of use cases, including image-based trait measurements, 3D reconstructions, morphological growth simulations, and crop modeling. Pseudo-batch/stream processing is also necessary; as data scales, manual batch jobs rapidly become infeasible, and (re-)analysis must occur upon arrival in near-real-time. We suggest web-accessible phenotyping automation software may address bottlenecks and help reveal undiscovered relationships between genes, traits, and the environment.

Cover crops may influence soil health and functioning. However, little is known about the role of belowground root architectural traits in linking cover crop diversity with rhizosphere soil ecosystem properties. We hypothesize that cover crop diversity may improve root traits, which in return, could influence its effects on essential indicators of soil physicochemical heterogeneity, such as the composition of soil aggregate-size classes and nutrients, the soil organic matter (SOM) and soil organic carbon (SOC) contents, and microbial communities. We studied the impact of cover plant diversity on root traits, soil properties and microbial communities. The four soil aggregate-size classes, such as large macro- (>2000μm), small macro- (<2000-500μm), meso- (<500-250 μm), and micro-aggregates (<250 μm) were separated by the dry sieving. Root traits such as surface area (cm2) and length (cm) were quantified by image analysis using Winrhizo. The soil nutrient, SOM, and SOC contents were determined by standard methods. Plant diversity improved productivity, root architectural traits, composition of soil aggregate-size classes and nutrients, SOM and SOC contents, composition and networking of microbial communities. Our results suggest that competition among plant roots in species-rich than poor communities may improve rhizosphere soil carbon storage, composition of soil aggregate-size classes and microbial communities.

Crop leaf angle is a crucial feature of plant architecture which influences photosynthetic efficiency and yield. Therefore high-throughput mapping of leaf angle is of extreme interest for both precision agriculture and crop breeding operations. In this study, we propose a UAV-based hybrid approach by combining a radiative transfer model (PROSAIL) and deep neural networks to estimate leaf angle from multi-angular hyperspectral and LiDAR data. PROSAIL can simulate canopy hyperspectral reflectance from a given list of parameters, where Average Leaf Inclination Angle (ALIA) is one of the canopy parameters. The goal is to develop a deep learning-based inversion function that takes UAV hyperspectral reflectance and other PROSAIL parameters as input and estimate ALIA of Maize as output. The other PROSAIL parameters will be estimated using several machine learning pipelines from UAV hyperspectral and LiDAR information. We also propose a multi-angular reflectance scheme where each image pixel will generate multiple simulated and observed reflectance from the overlapping regions using different angles (i.e., solar zenith angle, viewing zenith angle, and relative azimuth angle between the sun and sensor) involved in the PROSAIL simulation. An automated Python-based tool was developed that can calculate all three PROSAIL angles for a given hyperspectral data cube and generate the simulated reflectance for every vegetation pixels per experimental plot. Since the proposed method incorporates both crop information (i.e., PROSAIL) and a data-driven approach (i.e., deep learning), the method can be easily transferable for other study areas and crops, and it can rely on less ground truth data.

Predicting the composition of soybean seeds while the plants are growing in the field is very important to understand how different genotypes, field condition and environment influence different seed composition parameters. Knowing this information at global scale is even more important to understand the dynamics of food insecurity and the interaction of seed composition with global environmental changes. This study aims to develop a machine learning-based soybean seed composition model from the fusion of PlanetScope, Sentinel and Landsat satellite images. Although satellite images provide global coverage throughout the year, it suffers from coarser spatial resolution. However, PlanetScope provides four-band (i.e., red, green, blue, and near infrared) multispectral imageries at approximately 3m spatial resolution daily. Alternatively, Sentinel-2B and Landsat-8 have coarser spatial resolution (10 - 30m), they provide enriched spectral resolution. Therefore, the objectives of this study are to 1) fuse the PlanetScope image with corresponding Landsat and Sentinel images, 2) evaluate several machine learning algorithms (e.g., partial least squares, support vector machine, random forest, and deep neural network) to predict protein and oil content of soybean seeds from the fused satellite images. Two soybean fields were established in 2020 and 2021 at Bradford, MO to perform the experiment. Corresponding PlanetScope, Sentinel, and Landsat images were downloaded and processed for the entire growth seasons. Current results indicate that deep neural network provide the best performance in predicting both protein and oil content of soybean. Future step is to assess different fusion algorithms and predict seed composition at regional or global scale.

Fusiform rust is a disease caused by the fungal pathogen Cronartium quercuum f. sp. fusiforme (Cqf). It is considered one of the most damaging and economically important diseases for loblolly pine (Pinus taeda L.), causing millions of dollars in damage and loss of products each year. Evaluating trees for disease resistance includes inoculation trials – these trials require artificial inoculation of the pathogen followed by visual inspection for disease incidence. Visual inspection can lead to incorrect classification due to human error or escaped susceptible (i.e. a susceptible individual with no symptoms). Here, we plan to use vibrational spectroscopy tools to improve the accuracy of phenotypic values. Vibrational spectroscopy tools allow for a user to obtain a single, comprehensive reading based on the chemical constituents of sample. Because pine trees mostly rely on chemical-based defenses, the relationship between chemical makeup and resistance is promising. We plan to collect spectra from 40 different loblolly pine families (20 with lower rust incidence and 20 higher rust incidence) over five different progeny test sites in the southeastern US, totaling 400 trees. We will use a handheld near-infrared (NIR) spectrometer for a real-time, in-field reading on phloem and needle tissue. In addition, phloem and needle tissue will be analyzed by a benchtop Fourier-transformed infrared (FT-IR) spectrometer. Using multivariate analyses and machine learning algorithms, spectral readings can be mined for patterns associated with fusiform rust disease resistance or susceptibility, which can be used to predict the phenotype of untested trees. The results of the two tools and two tissue types will be compared to evaluate the best method for identifying phenotype in the system. This chemical fingerprinting and classification approach to phenotyping loblolly pines will provide a more objective, efficient, and more accurate way to identify disease resistance in the field, thereby creating more robust forest stands against fusiform rust.

We quantify the shape of hooked hairs which is a newly observed phenotype of epidermal cell extensions [1] in the common bean genotype L88-57 (Phaseolus vulgaris). The hooked hairs emerge below-ground before the root hairs and have a distinct ‘hooking’ morphology. We generated a dataset capturing their full distribution under the microscope within 3-5 days of germination. We quantify their shape by a novel computational pipeline that can automatically phenotype morphology. Our phenotyping pipeline quantifies traits like length, curvature, perimeter, area, and ‘hooking.’ Our objective is to quantify the response of hooked hairs to nutrient-stress to determine their function in common bean during early development. We used the pipeline for analyzing our dataset of hydroponically grown beans and observed statistically significant responses compared to the control for length, curvature, perimeter, and area to nitrogen (p<0.001**) and phosphorus (p<0.001**) stress treatments. The calculation of ‘hooking’ for our dataset is still ongoing. We are simultaneously developing a landmark-free method for the two-dimensional shape analysis of our dataset and believe that our phenotyping efforts will enable the high-throughput analysis of morphological root hair traits for any plant species.

Shape is foundational to biology. Observing and documenting shape has fueled biological understanding as the shape of biomolecules, cells, tissues, and organisms arise from the effects of genetics, development, and the environment. The vision of Topological Data Analysis (TDA), that data is shape and shape is data, will be relevant as biology transitions into a data-driven era where meaningful interpretation of large datasets is a limiting factor. We focus on quantifying the morphology of X-ray CT scans of a diverse panel of citrus and their tissues. In total, we have access to 146 different citrus samples comprising 58 different species, including samples of all fundamental citrus species. First, using the power of X-rays and image processing, we are able to compare and contrast size ratios between different tissues, such as the size of exocarp compared to the mesocarp or the endocarp. Second, using Persistent Homology we focus on characterizing the size and distribution of oil glands across different citrus species, and the relation of such distribution to other phenotypes such as exocarp surface area. This shape characterization will allow us later to link genotype with phenotype, furthering our understanding on how the physical shape is genetically coded in DNA.

Anthropogenic factors such as climate change, harsh agricultural practices, and mining have contributed to increases in soil salinization and heavy metal contamination. Highly saline environments drastically lower yield for crop species and elevated levels of toxic metals like cadmium are carcinogenic in the environment. Some plants that have evolved in high-salinity habitats or in soils with heavy metals could be used to remediate contaminated soils. Halophytes are plants with various adaptations that allow them to survive and reproduce in saline conditions. General mechanisms for salt tolerance/uptake in halophytes are hypothesized to help deal with other stresses like heavy metals. Plants with these traits could be utilized to extract salt and heavy metals from affected soils in a process called phytoremediation. We plan to develop Sea Rocket (Cakile maritima) in the Mustard family as a model system to understand mechanisms of salt (NaCl) and cadmium uptake and tolerance, as it has been shown to accumulate both. As part of this study, we will hydroponically grow C. maritima in different stress treatments using salt and cadmium. As the plants uptake the pollutants, the conductivity of the solution will change. We will develop an automated pipeline to track these changes in real-time using conductivity sensors. In addition, we will sample root and leaf tissues at various time points to measure salt and cadmium uptake using ICP-OES elemental analysis. This data will provide insights into salt and cadmium uptake/tolerance and paves a path toward efficient and viable solutions improving phytoremediation approaches.

Emergence timing of a plant, i.e., the time at which the plant is first visible from the surface of the soil, is an important phenotypic event as an indicator of seed quality and plant growth. Uneven emergence timing is associated with lower yield and poor farmer acceptance. The research introduces a novel deep learning based method called EmergeNet with a custom-designed loss function for coleoptile emergence timing detection and tracking its growth from a time-lapse video sequence in presence of cluttered background and extreme variations in illumination. EmergeNet uses a novel ensemble technique that integrates SEResNet, InceptionV3 and VGG19 to detect the coleoptile at its first tiny appearance on the soil surface. Emergence is an important phenotype which not only helps determine the dormancy of seeds for different genotypes in different conditions but also helps determine various aspects of the plant growth at an early stage. To develop and evaluate the algorithm, a benchmark dataset is indispensable. Thus, we introduce and publicly release a benchmark dataset called University of Nebraska-Lincoln Maize Emergence Dataset (UNL-MED). A visible light camera was fitted to capture the top view time-lapse images to form UNL-MED, where imaging started before the emergence and continued until maize seedlings are about 1 inch tall. Experimental evaluation on UNL-MED demonstrates the efficacy of the EmergeNet to detect the emergence timing with 100% accuracy when compared with human perceived groundtruth.

ABSTRACT Counting maize tassels in field conditions is predominantly done manually. Recently, computer-vision based methods have been utilized to detect tassels from images captured by UAV transects or poled-mounted cameras [1], [2], [3]. Once tassels are detected, deep-learning based local regression methods, Tasselnet, have been used to estimate in-field tassel counts [4]. However, field images are mostly captured over a period of time. Consequently, the input images in the foregoing Tasselnet technique are not independent but often form unequal sequences of correlated images. As such, the temporal sequence of images offers information about the growth trajectory of the plants. We propose a hybrid model that (a) utilizes convolutional neural network-based tassel localization in images, and (b) drives the local count of tassels utilizing the plant growth trajectory learned from the time-series of images. The resulting model can also handle important auxiliary information, obtained from in-field sensors (for example: soil moisture, air temperature etc.), that impacts plant growth and tassel counts. We implement our methodology on benchmark dataset [4] and compare our results with the SOTA Tasselnet [5]. Our initial results suggest that our technique is computationally viable and can produce accurate point estimates of tassel counts along with interval estimates capturing the precision of our estimates. Keywords: Computer vision, Convolutional neural networks, Deep learning, Maize tassels, Time-series. REFERENCES [1] Shi, Y., Alzadjali, A., Alali, M., Veeranampalayam-Sivakumar, A. N., Deogun, J., Scott, S., & Schnable, J. (2021). Maize tassel detection from UAV imagery using deep learning. Dryad. https://doi.org/10.5061/dryad.r2280gbcg. [2] Mirnezami, S. V., Srinivasan, S., Zhou, Y., Schnable, P. S., & Ganapathysubramanian, B. (2021). Detection of the Progression of Anthesis in Field-Grown Maize Tassels: A Case Study. Plant Phenomics, 2021, 4238701. doi:10.34133/2021/4238701. [3] Shete, S., Srinivasan, S., & Gonsalves, T. A. (2020). TasselGAN: An Application of the Generative Adversarial Model for Creating Field-Based Maize Tassel Data. Plant Phenomics, 2020, 8309605. doi:10.34133/2020/8309605. [4] Lu, H., Cao, Z., Xiao, Y., Zhuang, B., & Shen, C. (2017). TasselNet: counting maize tassels in the wild via local counts regression network. Plant Methods, 13(1), 79. doi:10.1186/s13007-017-0224-0. [5] Xiong, H., Cao, Z., Lu, H., Madec, S., Liu, L., & Shen, C. (2019). TasselNetv2: in-field counting of wheat spikes with context-augmented local regression networks. Plant Methods, 15(1), 150. doi:10.1186/s13007-019-0537-2.

This presentation talks about the effect of portable ground control points (GCPs) on the accuracy of unmanned aerial vehicle (UAV)-based remote sensing data in predicting plant health. 6 GCPs equipped with GPS receivers were spaced around the experimental plots of citrus and strawberry. UAVs equipped with multispectral sensors were then used to collect the remote sensing data of citrus and strawberry plants. The remote sensing data was used to calculate various vegetation indices including normalized difference vegetation index (NDVI), Green NDVI, and soil adjusted vegetation indices (SAVI). These indices were compared with the data obtained from proximal sensors that include Handheld Spectroradiometer and Chlorophyll Meter. Correlation between various vegetation indices, chlorophyll content, and spectroradiometer data will be shown and discussed. A significantly higher correlation coefficients were obtained between the remote sensing and proximal sensor data when the GCPs were used. Increased accuracy of remote sensing data is important for the widespread adoption of UAV-based remote sensing technology for precision agriculture so that the technology can be used for site-specific input management by taking into account the infield variability.