Plant Methods

BackgroundRecent developments in hybridization chain reaction (HCR) have enabled robust simultaneous localization of multiple mRNA transcripts using fluorescence in situ hybridization (FISH). Once multiple split initiator oligonucleotide probes bind their target mRNA, HCR uses DNA base-pairing of fluorophore-labeled hairpin sets to self-assemble into large polymers, amplifying the fluorescence signal and reducing non-specific background. Few studies have applied HCR in plants, despite its demonstrated utility in whole mount animal tissues and cell culture. Our aim was to optimize this technique for sectioned plant tissues embedded with paraffin and methacrylate resins, and to test its utility in combination with immunolocalization and subsequent correlation with cell ultrastructure using scanning electron microscopy. ResultsApplication of HCR to 10 {micro}m paraffin sections of 17-day-old Setaria viridis (green millet) inflorescences using confocal microscopy revealed that the transcripts of the transcription factor KNOTTED 1 (KN1) were localized to developing floret meristem and vascular tissue while SHATTERING 1 (SH1) and MYB26 transcripts were co-localized to the breakpoint below the floral structures (the abscission zone). We also used methacrylate de-embedment with 1.5 {micro}m and 0.5 {micro}m sections of 3-day-old Arabidopsis thaliana seedlings to show tissue specific CHLOROPHYLL BINDING FACTOR a/b (CAB1) mRNA highly expressed in photosynthetic tissues and ELONGATION FACTOR 1 ALPHA (EF1) highly expressed in meristematic tissues of the shoot apex. The housekeeping gene ACTIN7 (ACT7) mRNA was more uniformly distributed with reduced signals using lattice structured-illumination microscopy. HCR using 1.5 {micro}m methacrylate sections was followed by backscattered imaging and scanning electron microscopy thus demonstrating the feasibility of correlating fluorescent localization with ultrastructure. ConclusionHCR was successfully adapted for use with both paraffin and methacrylate de-embedment on diverse plant tissues in two model organisms, allowing for concurrent cellular and subcellular localization of multiple mRNAs, antibodies and other affinity probe classes. The mild hybridization conditions used in HCR made it highly amenable to observe immunofluorescence in the same section. De-embedded semi-thin methacrylate sections with HCR were compatible with correlative electron microscopy approaches. Our protocol provides numerous practical tips for successful HCR and affinity probe labeling in electron microscopy-compatible, sectioned plant material.

Macroautophagy is frequently quantified by live imaging of autophagosomes decorated with a marker of fluorescently tagged ATG8 protein (FT-ATG8) in Arabidopsis thaliana. This requires generation of suitable plant material by time-consuming crossing or transformation with FT-ATG8 marker. Autophagosome quantification by image analysis often relies on their counting in individual focal planes. This approach is prone to deliver biased results due to inappropriate sampling of the regions of interest in the Z-direction, as the actual 3D distribution of autophagosomes is usually not taken into account. To overcome such drawbacks, we have developed and tested a workflow consisting of immunofluorescence microscopy of autophagosomes labelled with anti-ATG8 antibody followed by stereological image analysis employing the optical disector and the Cavalieri principle. Our immunolabelling protocol specifically recognized autophagosomes in epidermal cells of A. thaliana root. Higher numbers of immunolabelled autophagosomes were observed when compared with those recognized with FT-AtATG8e marker, suggesting that single AtATG8 isoform markers cannot detect all autophagosomes in a cell. Therefore, immunolabelling provides more precise information as the anti-ATG8 antibody recognizes virtually all AtATG8 isoforms. The number of autophagosomes per tissue volume determined by stereological methods correlated with the intensity of autophagy induction treatment. Compared to autophagosome quantifications in maximum intensity projections, stereological methods detected autophagosomes present in a given volume with higher accuracy. Our novel application of immunolabelling combined with stereological methods constitutes a powerful toolbox for unbiased and reproducible quantification of autophagosomes and offers a convenient alternative to the standard of live imaging using FP-ATG8 marker.

Accurate 3D reconstruction is essential for high-throughput plant phenotyping, particularly for studying complex structures such as root systems. While photogrammetry and Structure from Motion (SfM) techniques have become widely used for 3D root imaging, the camera settings used are often underreported in studies, and the impact of camera calibration on model accuracy remains largely underexplored in plant science. In this study, we systematically evaluate the effects of focus, aperture, exposure time, and gain settings on the quality of 3D root models made with a multi-camera scanning system. We show through a series of experiments that calibration significantly improves model quality, with focus misalignment and shallow depth of field (DoF) being the most important factors affecting reconstruction accuracy. Our results further show that proper calibration has a greater effect on reducing noise than filtering it during post-processing, emphasizing the importance of optimizing image acquisition rather than relying solely on computational corrections. This work improves the repeatability and accuracy of 3D root phenotyping by giving useful calibration guidelines. This leads to better trait quantification for use in crop research and plant breeding.

Root phenotyping is a challenging task that would require monitoring root growth in soil under dark conditions to mimic natural conditions, while allowing the shoot to grow in light. Most existing methods involve exposing the roots to light, which substantially alters their growth and function. In this paper, we present an improved imaging system that can overcome this limitation of experiments performed in laboratories. The Dynamic Dark Root imaging Chamber (DDrC) enables continuous monitoring and image acquisition to track the dynamic development of root architecture under controlled growth conditions. Our imaging system is based on a Raspberry Pi camera module and infrared LEDs, which do not induce any stress responses in the roots. The DDrC setup is simple, affordable, and suitable for dynamic phenotyping experiments. We provide a detailed tutorial for the assembly and adjustment of the imaging chamber. We conclude that our system is a valuable tool for studying the genetic and environmental factors that affect the root system architecture and development, and for identifying the root traits that are related to plant adaptation and performance.

Root systems of crops play a significant role in agro-ecosystems. The root system is essential for water and nutrient uptake, plant stability, symbiosis with microbes and a good soil structure. Minirhizotrons, consisting of transparent tubes that create windows into the soil, have shown to be effective to non-invasively investigate the root system. Root traits, like root length observed around the tubes of minirhizotron, can therefore be obtained throughout the crop growing season. Analyzing datasets from minirhizotrons using common manual annotation methods, with conventional software tools, are time consuming and labor intensive. Therefore, an objective method for high throughput image analysis that provides data for field root-phenotyping is necessary. In this study we developed a pipeline combining state-of-the-art software tools, using deep neural networks and automated feature extraction. This pipeline consists of two major components and was applied to large root image datasets from minirhizotrons. First, a segmentation by a neural network model, trained with a small image sample is performed. Training and segmentation are done using "Root-Painter". Then, an automated feature extraction from the segments is carried out by "RhizoVision Explorer". To validate the results of our automated analysis pipeline, a comparison of root length between manually annotated and automatically processed data was realized with more than 58,000 images. Mainly the results show a high correlation (R=0.81) between manually and automatically determined root lengths. With respect to the processing time, our new pipeline outperforms manual annotation by 98.1 - 99.6 %. Our pipeline,combining state-of-the-art software tools, significantly reduces the processing time for minirhizotron images. Thus, image analysis is no longer the bottle-neck in high-throughput phenotyping approaches.

In the past few years, there has been an explosion in single-cell transcriptomics datasets, yet in vivo confirmation of these datasets is hampered in plants due to lack of robust validation methods. Likewise, modeling of plant development is hampered by paucity of spatial gene expression data. RNA fluorescence in situ hybridization (FISH) enables investigation of gene expression in the context of tissue type. Despite development of FISH methods for plants, easy and reliable whole mount FISH protocols have not yet been reported. We adapt a 3-day whole mount RNA-FISH method for plant species based on a combination of prior protocols that employs hybridization chain reaction (HCR), which amplifies the probe signal in an antibody-free manner. Our whole mount HCR RNA-FISH method shows expected spatial signals with low background for gene transcripts with known spatial expression patterns in Arabidopsis inflorescences and monocot roots. It allows simultaneous detection of three transcripts in 3D. We also show that HCR RNA-FISH can be combined with endogenous fluorescent protein detection and with our improved immunohistochemistry (IHC) protocol. The whole mount HCR RNA-FISH and IHC methods allow easy investigation of 3D spatial gene expression patterns in entire plant tissues.

BackgroundStomata are tiny pores on the leaf surface that are central to gas exchange. Stomatal number, size and aperture are key determinants of plant transpiration and photosynthesis, and variation in these traits can affect plant growth and productivity. Current methods to screen for stomatal phenotypes are tedious and not high throughput. This impedes research on stomatal biology and hinders efforts to develop resilient crops with optimised stomatal patterning. We have developed a rapid non-destructive method to phenotype stomatal traits in four species: wheat, rice, tomato and Arabidopsis. ResultsThe method consists of two steps. The first is the non-destructive capture of images of the leaf surface from plants in their growing environment using a handheld microscope; a process which only takes a few seconds compared to minutes for other methods. The second is to analyse stomatal features using a machine learning model that automatically detects, counts and measures stomatal number, size and aperture. The accuracy of the machine learning model in detecting stomata ranged from 76% to 99%, depending on the species, with a high correlation between measures of number, size and aperture between measurements using the machine learning models and by measuring them manually. The rapid method was applied to quickly identify contrasting stomatal phenotypes. ConclusionsWe developed a method that combines rapid non-destructive imaging of leaf surfaces with automated image analysis. The method provides accurate data on stomatal features while significantly reducing time for data acquisition and analysis. It can be readily used to phenotype stomata in large populations in the field and in controlled environments.

Root anatomical phenotyping has become a demonstrably essential part of investigating root physiology and in acquiring a holistic understanding of plant development. However, accessible high throughput methods for root anatomical analysis are still lacking. Here, we present the Rapid Anatomics Tool (RAT), a novel, low-cost system for high throughput root anatomical imaging with a shallow learning curve for obtaining high quality images suitable for comparative analysis across a number of plant species. Its efficiency comes from combining blockface-like imaging and stain-free imaging using near-ultraviolet (nUV) autofluorescence utilising a combination of low-cost commercial equipment, readily available mechanical components, and custom designed and 3D printed tools. Using this system, we investigated the anatomy of mature tissue along the axis of wheat crown roots, revealing a tendency of reduction in vascular complexity (expressed through a reduction in metaxylem number, area, and mean area per metaxylem file) from the basal to the distal region of the root. This study highlights the importance of thorough sampling strategies for investigating root anatomy in relation to organ function and introduces an accessible, relatively high-throughput method to support such research.

Plant diseases can cause heavy yield losses in arable crops resulting in major economic losses. Effective early disease recognition is paramount for modern large-scale farming. Since plants can be infected with multiple concurrent pathogens, it is important to be able to distinguish and identify each disease to ensure appropriate treatments can be applied. Hyperspectral imaging is a state-of-the art computer vision approach, which can improve plant disease classification, by capturing a wide range of wavelengths before symptoms become visible to the naked eye. Whilst a lot of work has been done applying the technique to identifying single infections, to our knowledge, it has not been used to analyse multiple concurrent infections which presents both practical and scientific challenges. In this study, we investigated three wheat pathogens (yellow rust, mildew and Septoria), cultivating co-occurring infections, resulting in a dataset of 1,447 hyperspectral images of single and double infections on wheat leaves. We used this dataset to train four disease classification algorithms (based on four neural network architectures: Inception and EfficientNet with either a 2D or 3D convolutional layer input). The highest accuracy was achieved by EfficientNet with a 2D convolution input with 81% overall classification accuracy, including a 72% accuracy for detecting a combined infection of yellow rust and mildew. Moreover, we found that hyperspectral signatures of a pathogen depended on whether another pathogen was present, raising interesting questions about co-existence of several pathogens on one plant host. Our work demonstrates that the application of hyperspectral imaging and deep learning is promising for classification of multiple infections in wheat, even with a relatively small training dataset, and opens opportunities for further research in this area. However, the limited number of Septoria and yellow rust + Septoria samples highlights the need for larger, more balanced datasets in future studies to further validate and extend our findings under field conditions.

BackgroundAutomating stomatal trait measurement has gained popularity because of their inherent importance for field phenotyping application as stomata are critical for both carbon capture and water use efficiency in plants. Such tool has been reported for rice, wheat, tomato, barley and oil palm. However, none exist yet for canola, which is an important economic and agronomic crop globally. ResultsWe developed a new toolkit called Stomatal Comprehensive Automated Neural Network or SCAN by combining the use of high-resolution portable digital microscopy with machine learning based on You Only Look Once algorithm (YOLOv8). Digital micrographs of leaf surfaces enter the SCAN pipeline, which includes stomata detection, stomata segmentation and stomatal pore segmentation models, to measure stomatal density, stomatal size and stomatal pore area, respectively. In addition to SCANs ability to measure leaf stomatal traits in canola at 89 to 94% accuracy, we also showed that SCAN can be used to predict stomatal density even in species not included in the training set such as Arabidopsis, tobacco, rice, wheat, maize and proso millet. SCAN was designed for the biological science community with the premise that users are not required to possess advanced programming capabilities to manage dependency prerequisites, execute the models, and integrate the analysis. This was achieved by packaging the models into a desktop application system that can be accessed offline. ConclusionOverall, SCAN provides a non-destructive, real-time, portable, and high-throughput measurement of leaf stomatal traits in canola. The minimised hardware requirement and user-friendly desktop application system make SCAN suitable for field phenotyping application.

Photoautotrophs environmental responses have been extensively studied at the organism and ecosystem level. However, less is known about their photosynthesis at the single cell level. This information is needed to understand photosynthetic acclimation processes, as light changes as it penetrates cells, layers of cells or organs. Furthermore, cells within the same tissue may behave differently, being at different developmental/physiological stages. Here we describe a new approach for single-cell and subcellular photophysiology based on the customisation of confocal microscopy to assess chlorophyll fluorescence quenching by the saturation pulse method. We exploit this setup to: i. reassess the specialisation of photosynthetic activities in developing tissues of non-vascular plants; ii. identify a specific subpopulation of phytoplankton cells in marine photosymbiosis, which are consolidating metabolic connections with their animal hosts, and iii. testify to the link between light penetration and photoprotection responses inside the different tissues that constitute a plant leaf anatomy. MotivationVisualising photosynthetic responses in 3D is essential for understanding most acclimation processes, as light changes within photosynthetic tissues as it penetrates the absorbing/diffusing layers of the cells. To achieve this goal, we developed a new imaging workflow merging confocal microscopy and saturating pulse chlorophyll fluorescence detection. This method applies to samples characterised by increasing complexity and its simplicity will contribute to its widespread use in plant and microalgae photoacclimation studies.

Photosynthesis is an essential process in plants that synthesizes sugars used for growth and development, highlighting the importance of establishing robust methods to monitor photosynthetic activity. Infrared gas analysis (IRGA) can be used to track photosynthetic rates by measuring the CO2 assimilation/release from a plant. Although much progress has been made in the development of IRGA technologies, challenges remain when using this technique on small herbaceous plants such as Arabidopsis thaliana. The use of whole plant chambers can overcome the difficulties associated with applying bulky leaf clamps to small delicate leaves, however this introduces the risk of soil-based microorganisms skewing gas exchange measurements. Here, we present a simple method to efficiently perform IRGA on A. thaliana plants using a whole plant chamber that removes soil-borne effects from the measurements. We show that this method can be used to detect subtle changes in photosynthetic rates measured at different times of day, under different growth conditions, and between wild-type and plants with deficiencies in the photosynthetic machinery. Furthermore, we show that this method can be used to detect changes in photosynthetic rates even at very young developmental stages such as 10 d-old seedlings. This method contributes to the array of techniques currently used to perform IRGA on A. thaliana and can allow for the monitoring of photosynthetic rates of whole plants from young ages.

BackgroundGene expression analyses are conducted using multiple approaches and increasingly research has been focused on assessing gene expression at the level of a tissue or even single-cells. To date, methods to assess gene expression at the single-cell in plant tissues have been semi-quantitative, require tissue disruption, and/or involve laborious, possibly artifact-inducing manipulation. In this work, we used grape berries (Vitis vinifera L. Zinfandel) as a model in order to examine the validity and reproducibility of an in-situ gene expression analysis method combining a cell pressure probe (CPP) with quantitative PCR (qPCR).\n\nResultsWe developed a method to directly assess gene expression levels via qPCR from cellular fluids sampled in-situ with a CPP. Cellular fluids, with volumes in the picoliter range, were collected from intact berries with a CPP at various depths across skin and mesocarp tissues. The expression of a key anthocyanin biosynthetic gene, UDP-glucose: flavonoid 3-O-glucosyltransferase (VviUFGT), was analyzed as a test case since its expression is restricted to cells producing anthocyanins in grape berry skins during ripening. The method identifies samples contaminated with significant levels of genomic DNA by amplifying a region of VviUFGT that spans an intron. Therefore false positives were discarded which occurred in 28% of the samples tested. Shallow probing of skin cells showed high VviUFGT expression as expected while deeper probing of mesocarp cells resulted in no VviUFGT expression.\n\nConclusionsThe clear correspondence of VviUFGT expression to the targeted cell samples suggests that the in-situ gene expression analysis using a CPP is reliable and does not result in contamination as the probe moves through tissues. This method can be paired to single-cell transcriptomic analyses in the future. We conclude that this technique represents a minimally invasive method of sampling plant cells in-situ which creates an opportunity for the analysis of cellular level, spatiotemporal responses in heterogeneous plant tissues.

BackgroundIt has been proposed that engineering the C4 photosynthetic pathway into C3 crops could significantly increase yield. This goal requires an increase in the chloroplast compartment of bundle sheath cells in C3 species. To facilitate large-scale testing of candidate regulators of chloroplast development in the rice bundle sheath, a simple and robust method to phenotype this tissue in C3 species is required. ResultsWe established a leaf ablation method to accelerate phenotyping of rice bundle sheath cells. The approach allowed bundle sheath cell dimensions, chloroplast area and chloroplast number per cell to be measured. Using this method, bundle sheath cell dimensions of maize were also measured and compared with rice. Our data show that bundle sheath width but not length significantly differed between C3 rice and C4 maize. Comparison of paradermal versus transverse bundle sheath cell width indicated that bundle sheath cells were intact after leaf ablation. Moreover, comparisons of planar chloroplast areas and chloroplast numbers per bundle sheath cell between wild-type and transgenic rice lines expressing the maize GOLDEN-2 (ZmG2) showed that the leaf ablation method allowed differences in chloroplast parameters to be detected. ConclusionsLeaf ablation is a simple approach to accessing bundle sheath cell files in C3 species. We show that this method is suitable for obtaining parameters associated with bundle sheath cell size, chloroplast area and chloroplast number per cell.

When dealing with plant roots, a multi-scale description of the functional root structure is needed. Since the beginning of XXI century, new devices like laser confocal microscopes have been accessible for coarse root structure measurements, including 3D reconstruction. Most researchers are familiar with using simple 2D geometry visualization that does not allow quantitatively determination of key morphological features from an organ-like perspective. We provide here a detailed description of the quantitative methods available for three-dimensional (3D) analysis of root features at single cell resolution, including root asymmetry, lateral root analysis, xylem and phloem structure, cell cycle kinetics, and chromatin determination. Quantitative maps of the distal and proximal root meristems are shown for different species, including Arabidopsis thaliana, Nicotiana tabacum and Medicago sativa. A 3D analysis of the primary root tip showed divergence in chromatin organization and cell volume distribution between cell types and precisely mapped root zonation for each cell file. Detailed protocols are also provided. Possible pitfalls in the usage of the marker lines are discussed. Therefore, researchers who need to improve their quantitative root biology portfolio can use them as a reference.

Irrigation and fertilization stress in plants are limitations for securing global food production. Sustainable agriculture is at the heart of global goals because threats of a rapidly growing population and climate changes are affecting agricultural productivity. Plant phenotyping is defined as evaluating plant traits. Traditionally, this measurement is performed manually but with advanced technology and analysis, these traits can be observed automatically and nondestructively. A high correlation between plant traits, growth, biomass, and final yield has been found. From the early stages of plant development, lack of irrigation and fertilization directly influence developing stages, thus the final crop yield is significantly reduced. In order to evaluate drought and fertilization stress, plant height, as a morphological trait, is the most common one used in precision-agriculture research. The present study shows that three-dimension volumetric approaches are more representative markers for alerting growers to the early stages of stress in young banana plants for fine-scale phenotyping. This research demonstrates two different group conditions: 1) Normal conditions; and 2) zero irrigation and zero fertilization. The statistical analysis results show a successfully distinguished early stress with the volumetric traits providing new insights on identifying the key phenotypes and growth stages influenced by drought stress.

Despite recent progress in our understanding of the graft union formation, we still know little about the cellular events underlying the grafting process. This is partially due to the difficulty of reliably targeting the graft interface in electron microscopy to study its ultrastructure and three-dimensional architecture. To overcome this technological bottleneck, we developed a correlative light electron microscopy approach (CLEM) to study the graft interface with high ultrastructural resolution. Grafting hypocotyls of Arabidopsis thaliana lines expressing YFP or mRFP in the endoplasmic reticulum allowed the efficient targeting of the grafting interface for under light and electron microscopy. To explore the potential of our method to study sub-cellular events at the graft interface, we focused on the formation of secondary plasmodesmata (PD) between the grafted partners. We showed that 4 classes of PD were formed at the interface and that PD introgression into the call wall was initiated equally by both partners. Moreover, the success of PD formation appeared not systematic with a third of PD not spanning the cell wall entirely. Characterizing the ultrastructural characteristics of these failed PD gives us insights into the process of secondary PD biogenesis. We showed that the thinning of the cell wall and the endoplasmic reticulum-plasma membrane tethering seem to be required for the establishment of symplastic connections between the scion and the rootstock. The resolution reached in this work shows that our CLEM method offer a new scale to the study for biological processes requiring the combination of light and electron microscopy.

Advanced smartphone technology now integrates sophisticated sensors, increasing access to high-precision data acquisition. This study tested the hypothesis that the iPhone 13-Pro camera, with LiDAR technology, can accurately estimate maize leaf surface area (Zea mays). 3D point cloud models enabled non-destructive data collection, and four methods for canopy area extraction were evaluated in relation to plant transpiration rates. Results showed a strong correlation (R2=0.92, RMSE=49.78) between manually scanned and iPhone-estimated plant surface areas. Additionally, the stem-to-plant surface area ratio was found to be 12.3% (R2=0.9, RMSE=28.42). Using this ratio to predict canopy area showed a significant correlation (R2=0.83) with actual canopy measurements. The iPhones surface area measurement tool offers an advantage by scanning the entire plant surface, unlike traditional leaf area index measurements, which often cannot penetrate the canopy. Moreover, real-size surface measurement of the canopy correlated strongly (R2=0.83) with whole canopy transpiration rates measured gravimetrically. This study introduces a novel method for analyzing 3D plant traits using a portable, affordable, and accurate tool, which has the potential to enhance plant breeding and agricultural practices. 0. How to Use This TemplateThe template details the sections that can be used in a manuscript. Note that each section has a corresponding style, which can be found in the "Styles" menu of Word. Sections that are not mandatory are listed as such. The section titles given are for articles. Review papers and other article types have a more flexible structure. Remove this paragraph and start section numbering with 1. For any questions, please contact the editorial office of the journal or support@mdpi.com.

Single-cell technology is increasingly used to analyze the basis of molecular regulation and provide insights into different aspects of human diseases. Such technology is a breakthrough approach to study blood cancers by characterizing molecular information on a genome-wide scale at the single-cell level. These methods can be easily and successfully transferred to tracheomycotic plant pathogens, which cause host wilt. Ceratocystis platani is the causal agent of the Canker stain disease of plane tree (Platanus spp.), a lethal wilt disease spreading in Europe. To displace and separate different C. platani conidia types a dielectrophoretic approach was tested. The DNA of each conidium was isolated and analyzed and the target DNA was identified by a specific qPCR marker and by sequencing the amplicon. Our results showed that this technology is applicable to vascular plant pathogens. The fungal DNA was successfully extracted from single or pooled conidia and identified by both methods after whole genome amplification. The use of the single-cell technology will provide a new approach to the study of plant vascular diseases, allowing the study of single-spore molecular and physiological features not detectable in complex biological mixtures. Author summaryIn recent years, technologies for single-cell isolation have been developed in the study of human diseases, such as cancers, capable of obtaining genetic information at the single-cell level. In this study, these methods were transferred to a plant pathogen, Ceratocystis platani, which causes a lethal disease of plane tree. The single cell technique used was able to separate the different types of conidia of C. platani and analyze the DNA within each conidium. The use of single cell technology represents an important tool for the study of plant vascular diseases by allowing the study of molecular mechanisms that are difficult to detect in complex biological matrices.

High-throughput phenotyping is a modern technology to measure plant traits efficiently and in large scale by imaging systems over the whole growth season. Those images provide rich data for statistical analysis of plant phenotypes. We propose a pipeline to extract and analyze the plant traits for field phenotyping systems. The proposed pipeline include the following main steps: plant segmentation from field images, automatic calculation of plant traits from the segmented images, and functional curve fitting for the extracted traits. To deal with the challenging problem of plant segmentation for field images, we propose a novel approach on image pixel classification by transform domain neural network models, which utilizes plant pixels from greenhouse images to train a segmentation model for field images. Our results show the proposed procedure is able to accurately extract plant heights and is more stable than results from Amazon Turks, who manually measure plant heights from original images.