RESUMO
Understanding the relationships between local environmental conditions and plant structure and function is critical for both fundamental science and for improving the performance of crops in field settings. Wind-induced plant motion is important in most agricultural systems, yet the complexity of the field environment means that it remained understudied. Despite the ready availability of image sequences showing plant motion, the cultivation of crop plants in dense field stands makes it difficult to detect features and characterize their general movement traits. Here, we present a robust method for characterizing motion in field-grown wheat plants (Triticum aestivum) from time-ordered sequences of red, green, and blue images. A series of crops and augmentations was applied to a dataset of 290 collected and annotated images of ear tips to increase variation and resolution when training a convolutional neural network. This approach enables wheat ears to be detected in the field without the need for camera calibration or a fixed imaging position. Videos of wheat plants moving in the wind were also collected and split into their component frames. Ear tips were detected using the trained network, then tracked between frames using a probabilistic tracking algorithm to approximate movement. These data can be used to characterize key movement traits, such as periodicity, and obtain more detailed static plant properties to assess plant structure and function in the field. Automated data extraction may be possible for informing lodging models, breeding programs, and linking movement properties to canopy light distributions and dynamic light fluctuation.
Assuntos
Aprendizado Profundo , Triticum/fisiologia , Agricultura , Algoritmos , Cruzamento , Produtos Agrícolas , Meio Ambiente , Movimento (Física) , Fenótipo , VentoRESUMO
Three-dimensional (3D) computer-generated models of plants are urgently needed to support both phenotyping and simulation-based studies such as photosynthesis modeling. However, the construction of accurate 3D plant models is challenging, as plants are complex objects with an intricate leaf structure, often consisting of thin and highly reflective surfaces that vary in shape and size, forming dense, complex, crowded scenes. We address these issues within an image-based method by taking an active vision approach, one that investigates the scene to intelligently capture images, to image acquisition. Rather than use the same camera positions for all plants, our technique is to acquire the images needed to reconstruct the target plant, tuning camera placement to match the plant's individual structure. Our method also combines volumetric- and surface-based reconstruction methods and determines the necessary images based on the analysis of voxel clusters. We describe a fully automatic plant modeling/phenotyping cell (or module) comprising a six-axis robot and a high-precision turntable. By using a standard color camera, we overcome the difficulties associated with laser-based plant reconstruction methods. The 3D models produced are compared with those obtained from fixed cameras and evaluated by comparison with data obtained by x-ray microcomputed tomography across different plant structures. Our results show that our method is successful in improving the accuracy and quality of data obtained from a variety of plant types.
Assuntos
Imageamento Tridimensional/métodos , Modelos Anatômicos , Brotos de Planta/anatomia & histologia , Plantas/anatomia & histologia , Microtomografia por Raio-X/métodos , Algoritmos , Calibragem , Fenótipo , Folhas de Planta/anatomia & histologiaRESUMO
The water stress-associated hormone abscisic acid (ABA) acts through a well-defined signal transduction cascade to mediate downstream transcriptional events important for acclimation to stress. Although ABA signaling is known to function in specific tissues to regulate root growth, little is understood regarding the spatial pattern of ABA-mediated transcriptional regulation. Here, we describe the construction and evaluation of an ABSCISIC ACID RESPONSIVE ELEMENT (ABRE)-based synthetic promoter reporter that reveals the transcriptional response of tissues to different levels of exogenous ABA and stresses. Genome-scale yeast one-hybrid screens complemented these approaches and revealed how promoter sequence and architecture affect the recruitment of diverse transcription factors (TFs) to the ABRE. Our analysis also revealed ABA-independent activity of the ABRE-reporter under nonstress conditions, with expression being enriched at the quiescent center and stem cell niche. We show that the WUSCHEL RELATED HOMEOBOX5 and NAC DOMAIN PROTEIN13 TFs regulate QC/SCN expression of the ABRE reporter, which highlights the convergence of developmental and DNA-damage signaling pathways onto this cis-element in the absence of water stress. This work establishes a tool to study the spatial pattern of ABA-mediated transcriptional regulation and a repertoire of TF-ABRE interactions that contribute to the developmental and environmental control of gene expression in roots.
Assuntos
Ácido Abscísico/genética , Arabidopsis/genética , Regulação da Expressão Gênica de Plantas , Genes Reporter , Regiões Promotoras Genéticas , Ácido Abscísico/metabolismo , Ácido Abscísico/farmacologia , Arabidopsis/efeitos dos fármacos , Proteínas de Arabidopsis/genética , Proteínas de Arabidopsis/metabolismo , Dano ao DNA , Redes Reguladoras de Genes , Proteínas de Homeodomínio/genética , Proteínas de Homeodomínio/metabolismo , Plantas Geneticamente Modificadas , Elementos de Resposta , Transdução de Sinais/genética , Análise Espaço-Temporal , Estresse Fisiológico/genética , Fatores de Transcrição/genética , Fatores de Transcrição/metabolismo , Leveduras/genéticaRESUMO
Auxin is a key regulator of plant growth and development. Within the root tip, auxin distribution plays a crucial role specifying developmental zones and coordinating tropic responses. Determining how the organ-scale auxin pattern is regulated at the cellular scale is essential to understanding how these processes are controlled. In this study, we developed an auxin transport model based on actual root cell geometries and carrier subcellular localizations. We tested model predictions using the DII-VENUS auxin sensor in conjunction with state-of-the-art segmentation tools. Our study revealed that auxin efflux carriers alone cannot create the pattern of auxin distribution at the root tip and that AUX1/LAX influx carriers are also required. We observed that AUX1 in lateral root cap (LRC) and elongating epidermal cells greatly enhance auxin's shootward flux, with this flux being predominantly through the LRC, entering the epidermal cells only as they enter the elongation zone. We conclude that the nonpolar AUX1/LAX influx carriers control which tissues have high auxin levels, whereas the polar PIN carriers control the direction of auxin transport within these tissues.
Assuntos
Arabidopsis/metabolismo , Ácidos Indolacéticos/metabolismo , Raízes de Plantas/metabolismo , Transporte Biológico , Frações Subcelulares/metabolismoRESUMO
Background and Aims: Intercropping systems contain two or more species simultaneously in close proximity. Due to contrasting features of the component crops, quantification of the light environment and photosynthetic productivity is extremely difficult. However it is an essential component of productivity. Here, a low-tech but high-resolution method is presented that can be applied to single- and multi-species cropping systems to facilitate characterization of the light environment. Different row layouts of an intercrop consisting of Bambara groundnut ( Vigna subterranea ) and proso millet ( Panicum miliaceum ) have been used as an example and the new opportunities presented by this approach have been analysed. Methods: Three-dimensional plant reconstruction, based on stereo cameras, combined with ray tracing was implemented to explore the light environment within the Bambara groundnut-proso millet intercropping system and associated monocrops. Gas exchange data were used to predict the total carbon gain of each component crop. Key Results: The shading influence of the tall proso millet on the shorter Bambara groundnut results in a reduction in total canopy light interception and carbon gain. However, the increased leaf area index (LAI) of proso millet, higher photosynthetic potential due to the C4 pathway and sub-optimal photosynthetic acclimation of Bambara groundnut to shade means that increasing the number of rows of millet will lead to greater light interception and carbon gain per unit ground area, despite Bambara groundnut intercepting more light per unit leaf area. Conclusions: Three-dimensional reconstruction combined with ray tracing provides a novel, accurate method of exploring the light environment within an intercrop that does not require difficult measurements of light interception and data-intensive manual reconstruction, especially for such systems with inherently high spatial possibilities. It provides new opportunities for calculating potential productivity within multi-species cropping systems, enables the quantification of dynamic physiological differences between crops grown as monoculture and those within intercrops, and enables the prediction of new productive combinations of previously untested crops.
Assuntos
Produção Agrícola , Imageamento Tridimensional , Produção Agrícola/métodos , Imageamento Tridimensional/métodos , Luz , Modelos Teóricos , Panicum/crescimento & desenvolvimento , Fotossíntese , Vigna/crescimento & desenvolvimentoRESUMO
As multicellular organisms grow, positional information is continually needed to regulate the pattern in which cells are arranged. In the Arabidopsis root, most cell types are organized in a radially symmetric pattern; however, a symmetry-breaking event generates bisymmetric auxin and cytokinin signaling domains in the stele. Bidirectional cross-talk between the stele and the surrounding tissues involving a mobile transcription factor, SHORT ROOT (SHR), and mobile microRNA species also determines vascular pattern, but it is currently unclear how these signals integrate. We use a multicellular model to determine a minimal set of components necessary for maintaining a stable vascular pattern. Simulations perturbing the signaling network show that, in addition to the mutually inhibitory interaction between auxin and cytokinin, signaling through SHR, microRNA165/6, and PHABULOSA is required to maintain a stable bisymmetric pattern. We have verified this prediction by observing loss of bisymmetry in shr mutants. The model reveals the importance of several features of the network, namely the mutual degradation of microRNA165/6 and PHABULOSA and the existence of an additional negative regulator of cytokinin signaling. These components form a plausible mechanism capable of patterning vascular tissues in the absence of positional inputs provided by the transport of hormones from the shoot.
Assuntos
Arabidopsis/fisiologia , MicroRNAs/metabolismo , Modelos Biológicos , Reguladores de Crescimento de Plantas/metabolismo , Raízes de Plantas/crescimento & desenvolvimento , Feixe Vascular de Plantas/crescimento & desenvolvimento , Transdução de Sinais/fisiologia , Arabidopsis/metabolismo , Proteínas de Arabidopsis/metabolismo , Proteínas de Homeodomínio/metabolismo , Microscopia Confocal , Fatores de Transcrição/metabolismoRESUMO
The number of image analysis tools supporting the extraction of architectural features of root systems has increased in recent years. These tools offer a handy set of complementary facilities, yet it is widely accepted that none of these software tools is able to extract in an efficient way the growing array of static and dynamic features for different types of images and species. We describe the Root System Markup Language (RSML), which has been designed to overcome two major challenges: (1) to enable portability of root architecture data between different software tools in an easy and interoperable manner, allowing seamless collaborative work; and (2) to provide a standard format upon which to base central repositories that will soon arise following the expanding worldwide root phenotyping effort. RSML follows the XML standard to store two- or three-dimensional image metadata, plant and root properties and geometries, continuous functions along individual root paths, and a suite of annotations at the image, plant, or root scale at one or several time points. Plant ontologies are used to describe botanical entities that are relevant at the scale of root system architecture. An XML schema describes the features and constraints of RSML, and open-source packages have been developed in several languages (R, Excel, Java, Python, and C#) to enable researchers to integrate RSML files into popular research workflow.
Assuntos
Raízes de Plantas/anatomia & histologia , Linguagens de Programação , Software , Imageamento Tridimensional , Modelos Biológicos , Raízes de Plantas/crescimento & desenvolvimento , Raízes de Plantas/fisiologia , Fluxo de TrabalhoRESUMO
Photoinhibition reduces photosynthetic productivity; however, it is difficult to quantify accurately in complex canopies partly because of a lack of high-resolution structural data on plant canopy architecture, which determines complex fluctuations of light in space and time. Here, we evaluate the effects of photoinhibition on long-term carbon gain (over 1 d) in three different wheat (Triticum aestivum) lines, which are architecturally diverse. We use a unique method for accurate digital three-dimensional reconstruction of canopies growing in the field. The reconstruction method captures unique architectural differences between lines, such as leaf angle, curvature, and leaf density, thus providing a sensitive method of evaluating the productivity of actual canopy structures that previously were difficult or impossible to obtain. We show that complex data on light distribution can be automatically obtained without conventional manual measurements. We use a mathematical model of photosynthesis parameterized by field data consisting of chlorophyll fluorescence, light response curves of carbon dioxide assimilation, and manual confirmation of canopy architecture and light attenuation. Model simulations show that photoinhibition alone can result in substantial reduction in carbon gain, but this is highly dependent on exact canopy architecture and the diurnal dynamics of photoinhibition. The use of such highly realistic canopy reconstructions also allows us to conclude that even a moderate change in leaf angle in upper layers of the wheat canopy led to a large increase in the number of leaves in a severely light-limited state.
Assuntos
Carbono/metabolismo , Imageamento Tridimensional/métodos , Modelos Biológicos , Triticum/fisiologia , Fluorescência , Luz , Fotossíntese , Folhas de Planta/metabolismo , Folhas de Planta/fisiologiaRESUMO
Increased adoption of the systems approach to biological research has focused attention on the use of quantitative models of biological objects. This includes a need for realistic three-dimensional (3D) representations of plant shoots for quantification and modeling. Previous limitations in single-view or multiple-view stereo algorithms have led to a reliance on volumetric methods or expensive hardware to record plant structure. We present a fully automatic approach to image-based 3D plant reconstruction that can be achieved using a single low-cost camera. The reconstructed plants are represented as a series of small planar sections that together model the more complex architecture of the leaf surfaces. The boundary of each leaf patch is refined using the level-set method, optimizing the model based on image information, curvature constraints, and the position of neighboring surfaces. The reconstruction process makes few assumptions about the nature of the plant material being reconstructed and, as such, is applicable to a wide variety of plant species and topologies and can be extended to canopy-scale imaging. We demonstrate the effectiveness of our approach on data sets of wheat (Triticum aestivum) and rice (Oryza sativa) plants as well as a unique virtual data set that allows us to compute quantitative measures of reconstruction accuracy. The output is a 3D mesh structure that is suitable for modeling applications in a format that can be imported in the majority of 3D graphics and software packages.
Assuntos
Imageamento Tridimensional/métodos , Oryza/citologia , Triticum/citologia , Algoritmos , Modelos Teóricos , Oryza/crescimento & desenvolvimento , Folhas de Planta/citologia , Folhas de Planta/crescimento & desenvolvimento , Brotos de Planta/citologia , Brotos de Planta/crescimento & desenvolvimento , Software , Triticum/crescimento & desenvolvimentoRESUMO
Seedling root traits of wheat (Triticum aestivum L.) have been shown to be important for efficient establishment and linked to mature plant traits such as height and yield. A root phenotyping pipeline, consisting of a germination paper-based screen combined with image segmentation and analysis software, was developed and used to characterize seedling traits in 94 doubled haploid progeny derived from a cross between the winter wheat cultivars Rialto and Savannah. Field experiments were conducted to measure mature plant height, grain yield, and nitrogen (N) uptake in three sites over 2 years. In total, 29 quantitative trait loci (QTLs) for seedling root traits were identified. Two QTLs for grain yield and N uptake co-localize with root QTLs on chromosomes 2B and 7D, respectively. Of the 29 root QTLs identified, 11 were found to co-localize on 6D, with four of these achieving highly significant logarithm of odds scores (>20). These results suggest the presence of a major-effect gene regulating seedling root vigour/growth on chromosome 6D.
Assuntos
Raízes de Plantas/crescimento & desenvolvimento , Poliploidia , Locos de Características Quantitativas/genética , Plântula/crescimento & desenvolvimento , Plântula/genética , Triticum/crescimento & desenvolvimento , Triticum/genética , Cromossomos de Plantas/genética , Nitrogênio/metabolismo , Fenótipo , Raízes de Plantas/anatomia & histologia , Raízes de Plantas/genética , Característica Quantitativa HerdávelRESUMO
It is increasingly important in life sciences that many cell-scale and tissue-scale measurements are quantified from confocal microscope images. However, extracting and analyzing large-scale confocal image data sets represents a major bottleneck for researchers. To aid this process, CellSeT software has been developed, which utilizes tissue-scale structure to help segment individual cells. We provide examples of how the CellSeT software can be used to quantify fluorescence of hormone-responsive nuclear reporters, determine membrane protein polarity, extract cell and tissue geometry for use in later modeling, and take many additional biologically relevant measures using an extensible plug-in toolset. Application of CellSeT promises to remove subjectivity from the resulting data sets and facilitate higher-throughput, quantitative approaches to plant cell research.
Assuntos
Arabidopsis/citologia , Processamento de Imagem Assistida por Computador/métodos , Microscopia Confocal/métodos , Células Vegetais/metabolismo , Software , Estatística como Assunto , Arabidopsis/metabolismo , Biomarcadores/metabolismo , Membrana Celular/metabolismo , Núcleo Celular/metabolismo , Fluorescência , Genes ReporterRESUMO
We present a novel image analysis tool that allows the semiautomated quantification of complex root system architectures in a range of plant species grown and imaged in a variety of ways. The automatic component of RootNav takes a top-down approach, utilizing the powerful expectation maximization classification algorithm to examine regions of the input image, calculating the likelihood that given pixels correspond to roots. This information is used as the basis for an optimization approach to root detection and quantification, which effectively fits a root model to the image data. The resulting user experience is akin to defining routes on a motorist's satellite navigation system: RootNav makes an initial optimized estimate of paths from the seed point to root apices, and the user is able to easily and intuitively refine the results using a visual approach. The proposed method is evaluated on winter wheat (Triticum aestivum) images (and demonstrated on Arabidopsis [Arabidopsis thaliana], Brassica napus, and rice [Oryza sativa]), and results are compared with manual analysis. Four exemplar traits are calculated and show clear illustrative differences between some of the wheat accessions. RootNav, however, provides the structural information needed to support extraction of a wider variety of biologically relevant measures. A separate viewer tool is provided to recover a rich set of architectural traits from RootNav's core representation.
Assuntos
Processamento de Imagem Assistida por Computador/métodos , Raízes de Plantas/anatomia & histologia , Software , Algoritmos , Arabidopsis/anatomia & histologia , Brassica napus/anatomia & histologia , Meristema/anatomia & histologia , Oryza/anatomia & histologia , Raízes de Plantas/fisiologia , Sementes/crescimento & desenvolvimento , Triticum/anatomia & histologia , Interface Usuário-ComputadorRESUMO
Raman spectroscopy is a rapid method for analysing the molecular composition of biological material. However, noise contamination in the spectral data necessitates careful pre-processing prior to analysis. Here we propose an end-to-end Convolutional Neural Network to automatically learn an optimal combination of pre-processing strategies, for the classification of Raman spectra of superficial and deep layers of cartilage harvested from 45 Osteoarthritis and 19 Osteoporosis (Healthy controls) patients. Using 6-fold cross-validation, the Multi-Convolutional Neural Network achieves comparable or improved classification accuracy against the best-performing Convolutional Neural Network applied to either the raw or pre-processed spectra. We utilised Integrated Gradients to identify the contributing features (Raman signatures) in the network decision process, showing they are biologically relevant. Using these features, we compared Artificial Neural Networks, Decision Trees and Support Vector Machines for the feature selection task. Results show that training on fewer than 3 and 300 features, respectively, for the disease classification and layer assignment task provide performance comparable to the best-performing CNN-based network applied to the full dataset. Our approach, incorporating multi-channel input and Integrated Gradients, can potentially facilitate the clinical translation of Raman spectroscopy-based diagnosis without the need for laborious manual pre-processing and feature selection.
Assuntos
Aprendizado Profundo , Redes Neurais de Computação , Osteoartrite , Análise Espectral Raman , Humanos , Análise Espectral Raman/métodos , Osteoartrite/classificação , Osteoartrite/diagnóstico , Feminino , Masculino , Cartilagem Articular/patologia , Pessoa de Meia-Idade , Idoso , Osteoporose/diagnóstico , Máquina de Vetores de SuporteRESUMO
Supervised machine learning techniques are increasingly being combined with ultrasonic sensor measurements owing to their strong performance. These techniques also offer advantages over calibration procedures of more complex fitting, improved generalisation, reduced development time, ability for continuous retraining, and the correlation of sensor data to important process information. However, their implementation requires expertise to extract and select appropriate features from the sensor measurements as model inputs, select the type of machine learning algorithm to use, and find a suitable set of model hyperparameters. The aim of this article is to facilitate implementation of machine learning techniques in combination with ultrasonic measurements for in-line and on-line monitoring of industrial processes and other similar applications. The article first reviews the use of ultrasonic sensors for monitoring processes, before reviewing the combination of ultrasonic measurements and machine learning. We include literature from other sectors such as structural health monitoring. This review covers feature extraction, feature selection, algorithm choice, hyperparameter selection, data augmentation, domain adaptation, semi-supervised learning and machine learning interpretability. Finally, recommendations for applying machine learning to the reviewed processes are made.
Assuntos
Aprendizado de Máquina , Ultrassom , Algoritmos , Monitorização FisiológicaRESUMO
Chlorinated ethene (CE) groundwater contamination is commonly treated through anaerobic biodegradation (i.e., reductive dechlorination) either as part of an engineered system or through natural attenuation. Aerobic biodegradation has also been recognized as a potentially significant pathway for the removal of the lower CEs cis-1,2-dichloroethene (cDCE) and vinyl chloride (VC). However, the role of aerobic biodegradation under low oxygen conditions typical of contaminated groundwater is unclear. Bacteria capable of aerobic VC biodegradation appear to be common in the environment, while aerobic biodegradation of cDCE is less common and little is known regarding the organisms responsible. In this study, we investigate the role of aerobic cDCE and VC biodegradation in a mixed contaminant plume (including CEs, BTEX, and ketones) at Naval Air Station North Island, Installation Restoration Site 9. Sediment and groundwater collected from the plume source area, mid-plume, and shoreline were used to prepare microcosms under fully aerobic (8 mg/L dissolved oxygen (DO)) and suboxic (< 1 mg/L DO) conditions. In the shoreline microcosms, VC and cDCE were rapidly degraded under suboxic conditions (100% and 77% removal in < 62 days). In the suboxic VC microcosms, biodegradation was associated with a > 5 order of magnitude increase in the abundance of functional gene etnE, part of the aerobic VC utilization pathway. VC and cDCE were degraded more slowly under fully aerobic conditions (74% and 30% removal) in 110 days. High-throughput 16S rRNA and etnE sequencing suggest the presence of novel VC- and cDCE-degrading bacteria. These results suggest that natural aerobic biodegradation of cDCE and VC is occurring at the site and provide new evidence that low (< 1 mg/L) DO levels play a significant role in natural attenuation of cDCE and VC.
Assuntos
Água Subterrânea , Cloreto de Vinil , Poluentes Químicos da Água , Bactérias/metabolismo , Biodegradação Ambiental , Água Subterrânea/microbiologia , Oxigênio/metabolismo , RNA Ribossômico 16S/genética , Cloreto de Vinil/metabolismo , Poluentes Químicos da Água/metabolismoRESUMO
A combination of batch and column experiments evaluated the mass transfer of two candidate partitioning electron donors (PEDs), n-hexanol (nHex) and n-butyl acetate (nBA), for enhanced bioremediation of trichloroethene (TCE)-dense nonaqueous phase liquid (DNAPL). Completely mixed batch reactor experiments yielded equilibrium TCE-DNAPL and water partition coefficients (KNW) for nHex and nBA of 21.7 ± 0.27 and 330.43 ± 6.7, respectively, over a range of initial PED concentrations up to the aqueous solubility limit of ca. 5000 mg/L. First-order liquid-liquid mass transfer rates determined in batch reactors with nBA or nHex concentrations near the aqueous solubility were 0.22 min(-1) and 0.11 min(-1), respectively. Liquid-liquid mass transfer under dynamic flow conditions was assessed in one-dimensional (1-D) abiotic columns packed with Federal Fine Ottawa sand containing a uniform distribution of residual TCE-DNAPL. Following pulse injection of PED solutions at pore-water velocities (vp) ranging from 1.2 to 6.0 m/day, effluent concentration measurements demonstrated that both nHex and nBA partitioned strongly into residual TCE-DNAPL with maximum effluent levels not exceeding 35% and 7%, respectively, of the applied concentrations of 4000 to 5000 mg/L. PEDs persisted at effluent concentrations above 5 mg/L for up to 16 and 80 pore volumes for nHex and nBA, respectively. Mathematical simulations yielded KNW values ranging from 44.7 to 48.2 and 247 to 291 and liquid-liquid mass transfer rates of 0.01 to 0.03 min(-1) and 0.001 to 0.006 min(-1) for nHex and nBA, respectively. The observed TCE-DNAPL and water mass transfer behavior suggests that a single PED injection can persist in a treated source zone for prolonged time periods, thereby reducing the need for, or frequency of, repeated electron donor injections to support bacteria that derive reducing equivalents for TCE reductive dechlorination from PED fermentation.
Assuntos
Tricloroetileno/química , Acetatos/química , Biodegradação Ambiental , Compostos Clorados , Elétrons , Fermentação , Hexanóis/química , Modelos Teóricos , Solventes , Tricloroetileno/metabolismoRESUMO
3D reconstruction of fruit is important as a key component of fruit grading and an important part of many size estimation pipelines. Like many computer vision challenges, the 3D reconstruction task suffers from a lack of readily available training data in most domains, with methods typically depending on large datasets of high-quality image-model pairs. In this paper, we propose an unsupervised domain-adaptation approach to 3D reconstruction where labelled images only exist in our source synthetic domain, and training is supplemented with different unlabelled datasets from the target real domain. We approach the problem of 3D reconstruction using volumetric regression and produce a training set of 25,000 pairs of images and volumes using hand-crafted 3D models of bananas rendered in a 3D modelling environment (Blender). Each image is then enhanced by a GAN to more closely match the domain of photographs of real images by introducing a volumetric consistency loss, improving performance of 3D reconstruction on real images. Our solution harnesses the cost benefits of synthetic data while still maintaining good performance on real world images. We focus this work on the task of 3D banana reconstruction from a single image, representing a common task in plant phenotyping, but this approach is general and may be adapted to any 3D reconstruction task including other plant species and organs.
RESUMO
We address the complex problem of reliably segmenting root structure from soil in X-ray Computed Tomography (CT) images. We utilise a deep learning approach, and propose a state-of-the-art multi-resolution architecture based on encoderdecoders. While previous work in encoder-decoders implies the use of multiple resolutions simply by downsampling and upsampling images, we make this process explicit, with branches of the network tasked separately with obtaining local high-resolution segmentation, and wider low-resolution contextual information. The complete network is a memory efficient implementation that is still able to resolve small root detail in large volumetric images. We compare against a number of different encoder-decoder based architectures from the literature, as well as a popular existing image analysis tool designed for root CT segmentation. We show qualitatively and quantitatively that a multi-resolution approach offers substantial accuracy improvements over a both a small receptive field size in a deep network, or a larger receptive field in a shallower network. We then further improve performance using an incremental learning approach, in which failures in the original network are used to generate harder negative training examples. Our proposed method requires no user interaction, is fully automatic, and identifies large and fine root material throughout the whole volume.
RESUMO
Understanding plant growth processes is important for many aspects of biology and food security. Automating the observations of plant development-a process referred to as plant phenotyping-is increasingly important in the plant sciences, and is often a bottleneck. Automated tools are required to analyze the data in microscopy images depicting plant growth, either locating or counting regions of cellular features in images. In this paper, we present to the plant community an introduction to and exploration of two machine learning approaches to address the problem of marker localization in confocal microscopy. First, a comparative study is conducted on the classification accuracy of common conventional machine learning algorithms, as a means to highlight challenges with these methods. Second, a 3D (volumetric) deep learning approach is developed and presented, including consideration of appropriate loss functions and training data. A qualitative and quantitative analysis of all the results produced is performed. Evaluation of all approaches is performed on an unseen time-series sequence comprising several individual 3D volumes, capturing plant growth. The comparative analysis shows that the deep learning approach produces more accurate and robust results than traditional machine learning. To accompany the paper, we are releasing the 4D point annotation tool used to generate the annotations, in the form of a plugin for the popular ImageJ (FIJI) software. Network models and example datasets will also be available online.
RESUMO
The phytohormones salicylic acid (SA), jasmonic acid (JA), and ethylene (ET) are central regulators of biotic and abiotic stress responses in Arabidopsis thaliana. Here, we generated modular fluorescent protein-based reporter lines termed COLORFUL-PR1pro, -VSP2pro, and -PDF1.2apro. These feature hormone-controlled nucleus-targeted transcriptional output sensors and the simultaneous constitutive expression of spectrally separated nuclear reference and plasma membrane-localized reporters. This set-up allowed the study of cell-type specific hormone activities, cellular viability and microbial invasion. Moreover, we developed a software-supported high-throughput confocal microscopy imaging protocol for output quantification to resolve the spatio-temporal dynamics of respective hormonal signaling activities at single-cell resolution. Proof-of-principle analyses in A. thaliana leaves revealed distinguished hormone sensitivities in mesophyll, epidermal pavement and stomatal guard cells, suggesting cell type-specific regulatory protein activities. In plant-microbe interaction studies, we found that virulent and avirulent Hyaloperonospora arabidopsidis (Hpa) isolates exhibit different invasion dynamics and induce spatio-temporally distinct hormonal activity signatures. On the cellular level, these hormone-controlled reporter signatures demarcate the nascent sites of Hpa entry and progression, and highlight initiation, transduction and local containment of immune signals.