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.

Paramecium cells swim and feed by beating their thousands of cilia in coordinated patterns. The organization of these patterns and its relationship with cell motility has been the subject of a large body of work, particularly as a model for ciliary beating in human organs where similar organization is seen. However the rapid motion of the cells makes quantitative measurements very challenging. Here we provide detailed measurements of the swimming of Paramecium cells from high-speed video at high magnification, as they move in microfluidic channels. An image analysis protocol allows us to decouple the cell movement from the motion of the cilia, thus allowing us to measure the ciliary beat frequency (CBF) and the spatio-temporal organization into metachronal waves along the cell periphery. Two distinct values of the CBF appear at different regions of the cell: most of the cilia beat in the range of 15 to 45 Hz, while the cilia in the peristomal region beat at almost double the frequency. The body and peristomal CBF display a nearly linear relation with the swimming velocity. Moreover the measurements do not display a measurable correlation between the swimming velocity and the metachronal wave velocity on the cell periphery. These measurements are repeated for four RNAi silenced mutants, where proteins specific to the cilia or to their connection to the cell base are depleted. We find that the mutants whose ciliary structure is affected display similar swimming to the control cells albeit with a reduced efficiency, while the mutations that affect the cilia's anchoring to the cell lead to strongly reduced ability to swim. This reduction in motility can be related to a loss of coordination between the ciliary beating in different parts of the cell.

Wind has major effect on plants, from growth changes to windbreaks. Therefore, there is a crucial need for non-invasive methods to describe and quantify the complex motion of a plant induced by wind. In this paper two methods based on video sequences analysis are studied. An adaptation of the classical Particle Image Velocimetry method (nat-PIV) is compared with a tracking method based on the optical flow method of Lukas and Kanade, initialized with the features selection method of Shi and Tomasi (ST + KLT). Both methods were benchmarked on an experiment on a walnut tree in open-field conditions submitted to different wind flows at different periods of the year and equipped with 3D magnetic tracking. The metrological assessment was performed in two steps. We first tested if the results given by both methods were significantly different. Secondly, a direct assessment of the two methods versus 3D magnetic tracking was performed. The ST + KLT method proved to be more accurate and robust than nat-PIV one. The outputs of the ST + KLT method are independent of the foliage density, wind velocity and of light gradient intrinsic to outdoor scene. The implementation of ST + KLT method developed for this study in Matlab is freely available. (C) 2013 Elsevier B.V. All rights reserved.

Model plants are extensively used in biological studies, and their mechanical behaviour needs to be better understood, in relation to studies in mechanoperception for instance. We present here the first approach to derive experimentally the modal parameters of two of these plants, Arabidopsis thaliana and Populus tremula × alba. A classical sinusoidal sweep excitation is used, with a measurement of displacements based on LKT optical flow tracking, followed by a bi-orthogonal decomposition (BOD). This allows us to estimate several modal frequencies for each plant, as well as the corresponding spatial localizations of deformation. Analyzing the modal frequencies, we show that global and local modes correspond to distinct ranges of frequencies and depend differently on plant size. Possible phenotyping applications are then discussed.

FSPM analysis of root systems requires structural data obtained on large data set. This paper describes a processing pipeline developed to extract automatically the architecture of root system from images databases.

Background and Aims Automatic acquisition of plant architecture is a major challenge for the construction of quantitative models of plant development. Recently, 3-D laser scanners have made it possible to acquire 3-D images representing a sampling of an object's surface. A number of specific methods have been proposed to reconstruct plausible branching structures from this new type of data, but critical questions remain regarding their suitability and accuracy before they can be fully exploited for use in biological applications.Methods In this paper, an evaluation framework to assess the accuracy of tree reconstructions is presented. The use of this framework is illustrated on a selection of laser scans of trees. Scanned data were manipulated by experienced researchers to produce reference tree reconstructions against which comparisons could be made. The evaluation framework is given two tree structures and compares both their elements and their topological organization. Similar elements are identified based on geometric criteria using an optimization algorithm. The organization of these elements is then compared and their similarity quantified. From these analyses, two indices of geometrical and structural similarities are defined, and the automatic reconstructions can thus be compared with the reference structures in order to assess their accuracy.Key Results The evaluation framework that was developed was successful at capturing the variation in similarities between two structures as different levels of noise were introduced. The framework was used to compare three different reconstruction methods taken from the literature, and allowed sensitive parameters of each one to be determined. The framework was also generalized for the evaluation of root reconstruction from 2-D images and demonstrated its sensitivity to higher architectural complexity of structure which was not detected with a global evaluation criterion.Conclusions The evaluation framework presented quantifies geometric and structural similarities between two structures. It can be applied to the characterization and comparison of automatic reconstructions of plant structures from laser scanner data and 2-D images. As such, it can be used as a reference test for comparing and assessing reconstruction procedures.

Vegetation is present all around us and we are accustomed to see trees and other plants everyday. Its accurate representation is thus an essential part of the realistic depiction of natural scenes in a virtual environment. Due to the complexity of both vegetation and its reaction to wind load, such representation are still being unresolved issues of research in computer graphics. In a first step to the development of suitable models of animated plants, it is thus essential to understand as much as possible all the phenomena which produce the observable motion of trees. This thesis is organized in two parts. The first is on the acquisition and the reproduction of the motion of real plants and contains three chapters. The second is on mechanical simulation of plants dynamics and is divided in two chapters. In the first part, chapter 2 describes experimental work I have done and participated to within the Chene-Roseau project and shows the obtained data on plants response to natural wind load and manual excitation. We also discuss how features observable in a video can be automatically tracked along the sequence and presents some methods I have developed. These algorithms have been put in a software I developed that also focus on user interaction to compensate for the limitation of automatic techniques. An explanation is given in chapter 3 of how manual input can be used to extract reliable motion data from video. At the end of the first part, chapter 4 presents our results on structure extraction from 2D motion data (previously extracted from a video) and its retargeting. The statistical study of 2D motion data I have developed is discussed. We show how it can be used to extract a valid hierarchical branches structure that holds the plants motion and how it is used to reproduce the observed motion on a virtual model. In the second part, simulation method of tree response to wind load is discussed. In chapter 5, a state of the art of existing real-time animation technique is given. We introduce several concept of mechanics and simulation of elastic structure dynamics in order to compare all described methods. Finally chapter 6 presents the methods developed in collaboration with Mathieu Rodriguez on the real-time simulation of thousands of trees in response to interactive wind.

This paper presents a real‐time method to animate complex scenes of thousands of trees under a user‐controllable wind load. Firstly, modal analysis is applied to extract the main modes of deformation from the mechanical model of a 3D tree. The novelty of our contribution is to precompute a new basis of the modal stress of the tree under wind load. At runtime, this basis allows to replace the modal projection of the external forces by a direct mapping for any directional wind. We show that this approach can be efficiently implemented on graphics hardware. This modal animation can be simulated at low computation cost even for large scenes containing thousands of trees.

The complexity of animating trees, shrubs and foliage is an impediment to the efficient and realistic depiction of natural environments. This paper presents an algorithm to extract, from a single video sequence, motion fields of real shrubs under the influence of wind, and to transfer this motion to the animation of complex, synthetic 3D plant models. The extracted motion is retargeted without requiring physical simulation. First, feature tracking is applied to the video footage, allowing the 2D position and velocity of automatically identified features to be clustered. A key contribution of the method is that the hierarchy obtained through statistical clustering can be used to synthesize a 2D hierarchical geometric structure of branches that terminates according to the cut-off threshold of a classification algorithm. This step extracts both the shape and the motion of a hierarchy of features groups that are identified as geometrical branches. The 2D hierarchy is then extended to three dimensions using the estimated spatial distribution of the features within each group. Another key contribution is that this 3D hierarchical structure can be efficiently used as a motion controller to animate any complex 3D model of similar but non-identical plants using a standard skinning algorithm. Thus, a single video source of a moving shrub becomes an input device for a large class of virtual shrubs. We illustrate the results on two examples of shrubs and one outdoor tree. Extensions to other outdoor plants are discussed.

La complexité de l'animation des arbres et de leur feuillageest un problème important en informatique graphique. Ce papier présente un algorithme qui permet à partir d'un cha mp de vitesse, extrait d'une vidéo de plante réelle animée par le vent, le transfert de son mouvement sur un modèl e de plante virtuelle 3D, et cela sans nécessiter l'utilisation de simulateur physique. Un ensemble de point s est suivi dans chacune des images de la vidéo puis est classifié d'après les variations de positions et de vitesses au cours de la séquence. Une contribution principale de cette méthode est que la hiérarchie obtenue par un méthodede classification est utilisée pour générer une structure géométrique 2D de branches. Cette hiérarchie 2D e st ensuite étendue en 3D grâce à une étude de la distribution spatiale des points suivis dans la vidéo. Une a utre contribution majeure est que cette structure 3D peut être utilisée, à l'aide d'un algorithme de skinning, comme contrôleur du mouvement pour animer un modèle virtuel de structure similaire mais pas nécessairement ide ntique. Nous exposons ici les résultats obtenus sur deux arbustes et un arbre. The complexity of animating trees, shrubs and foliage is an i mpediment to the efficient and realistic depiction of natural environments. This paper presents an algorithm to e xtract, from a single video sequence, the motion of real shrubs under the influence of wind, and to transfer this m otion to the animation of complex, synthetic 3D plant models. The extracted motion is retargeted without re quiring physical simulation. First, feature tracking is applied to the video footage, allowing the 2D position andvelocity of automatically identified features to be clustered. A key contribution of the method is that the hiera rchy obtained through statistical clustering can be used to synthesize a 2D hierarchical geometric structure ofbranches. This step extracts both the shape and the motion of a hierarchy of features groups. The 2D hierarchy isthen extended to 3D using the estimated spatial distribution of the features within each group. Another keycontribution is that this 3D hierarchical structure can be efficiently used as a motion controller to animate any comp lex 3D model of similar but non-identical plants using a standard skinning algorithm. We illustrate the resu lts on two examples of shrubs and discuss extensions to trees and other outdoor plants.