Portail HAL des publications du LIRMM
Not a member yet
13279 research outputs found
Sort by
Fully automatic extraction of morphological traits from the web: Utopia or reality?
No full textInternational audiencePlant morphological traits, their observable characteristics, are fundamental to understanding the role played by each species within its ecosystem; however, compiling trait information for even a moderate number of species is a demanding task that may take experts years to accomplish. At the same time, online species descriptions contain massive amounts of information about morphological traits, but the lack of structure makes this source of data impossible to use at scale. Methods: To overcome this, we propose to leverage recent advances in large language models and devise a mechanism for gathering and processing plant trait information in the form of unstructured textual descriptions, without manual curation. Results: We evaluate our approach by automatically replicating three manually created species-trait matrices. Our method found values for over half of all species-trait pairs, with an F1 score of over 75%. Discussion: Our results suggest that large-scale creation of structured trait databases from unstructured online text is now feasible due to the information extraction capabilities of large language models. However, the process is currently limited by the availability of textual descriptions that cover all traits of interest
Introducing locality in some generalized AG codes
No full textInternational audienceIn 1999, Xing, Niederreiter and Lam introduced a generalization of AG codes (GAG codes) using the evaluation at non-rational places of a function field. In this paper, we show that one can obtain a locality parameter r in such codes by using only non-rational places of degree at most r. This is, up to the author’s knowledge, a new way to construct locally recoverable codes (LRCs). We give an example of such a code reaching the Singleton-like bound for LRCs, and show the parameters obtained for some longer codes over . We then investigate similarities with some concatenated codes. Contrary to previous methods, our construction allows one to obtain directly codes whose dimension is not a multiple of the locality. Finally, we give an asymptotic study using the Garcia–Stichtenoth tower of function fields, for both our construction with GAG codes and a construction of concatenated codes. We give explicit infinite families of LRCs with locality 2 over any finite field of cardinality greater than 3 following our approach with GAG codes
Semi-automated analysis of cerebral capillary red blood cell velocities allows modeling of transit time distribution after experimental subarachnoid hemorrhage in mice
No full textInternational audienceSignificance: Microvascular dysfunction stems from the origin of various neurological diseases. Among these, delayed cerebral ischemia following subarachnoid hemorrhage (SAH) is a major complication. Even though pathogenesis remains poorly understood, hypotheses converge toward early and persistent microvascular dysfunction. In this context, mathematical models have been developed to study oxygen delivery using theoretical distributions of capillary flux. However, these distributions lack experimental validation.Aim: We propose experimental recording of capillary red blood cell (RBC) velocities in a superficial cortical microvascular network in a mouse model of SAH, testing theoretical transit time distributions and their implications on tissue oxygenation.Approach: We performed optical recording of RBC velocities. We propose a complete software, available on GitHub, for velocity semi-automated measurement. Experimental data were fitted with Gamma and Cauchy probability distribution functions (PDFs). Corresponding maximal oxygen metabolic rates ( CMROmax2) were computed.Results: Data showed that transit time distributions changed after SAH, such that they followed a Cauchy distribution. Corresponding CMROmax2 maps showed a malignant capillary heterogeneity state.Conclusions: We provide distributions of transit times in an SAH mouse model, allowing us to discuss PDF implications for maximal oxygen consumption
Toward the prediction of flowering date of apple trees from unmanned aerial vehicle (UAV) imagery
No full textSource Agritrop Cirad (https://agritrop.cirad.fr/613776/)International audienceFruit tree flowering and reproduction processes are crucial for production. Flowering time is currently impacted by global warming affecting orchard production. The screening of genetic resources could provide new keys for fruit tree adaptation. Proxy-detection has great potential to help characterize plant cultivars and select the best performing ones. In this context, the ambition of this project is to develop new methodologies for predicting flowering dates using aerial images at regular time intervals. For this, we used a core-collection of 241 apple French cultivars with four repetitions per cultivar, planted in 2014 in Montpellier (south of France). RGB and multispectral images were acquired on the orchard, using drone-borne sensors, resulting in orthomosaics of the entire orchard. For each type of sensor, four periods of acquisition in 2021 (between June and November) were used to characterize different foliage phenological stages. In parallel, expert notations of the flowering dates were carried out in the following spring (2022). The eight orthomosaics were split into patches representing individual trees. Different methods of machine and deep learning were tested to predict the flowering dates from the patches. First, NDVI and NDRE were determined from each patch and regression methods, such as Lasso, Random Forest, Gradient Boosting, were applied. A convolutional neural network (CNN), inspired from U-Net, was then tested. This CNN was tested on individual patches and on concatenated patches of the different months. Finally, a clustering method was used to identify flowering periods by grouping trees with similar flowering dates and a CNN was trained to predict the flowering period of each tree. This last approach proved to be the most accurate with an RMSE of approximately 10 days. This work illustrates the interest and limitation of deep learning on images for the phenotyping of traits representing genotypic behavior at different tree developmental stages
Guide des bonnes pratiques pour la gestion des données de la recherche en BioImagerie: Cas d'usage en microscopie photonique
No full textInternational audienceCe guide a été conçu pour répondre aux défis récents de la gestion des données de recherche dans le contexte de la science ouverte. Il s'appuie sur les réflexions interdisciplinaires menées au sein de réseaux et de groupes de travail de la MITI (GeDeM, RTmfm et DOREMITI) et d'instituts du CNRS, mettant en avant des pratiques FAIR (Faciles à trouver, Accessibles, Interopérables et Réutilisables). En prenant pour exemple la gestion des données en microscopie photonique, ce document propose un cas d'usage concret tout en exposant les notions fondamentales de la gestion FAIR. Le guide suit les étapes du cycle de vie des données, enrichi d'une phase initiale dédiée à la planification et à la préparation des projets. Cette approche met en lumière l'importance de chaque étape, depuis l'acquisition jusqu'à la publication, afin d'assurer la pérennité, la diffusion et la réutilisation des données au-delà de leur contexte initial. Non exhaustif, ce guide s'inscrit dans les efforts nationaux pour la science ouverte, offrant un accompagnement aux ingénieurs et aux chercheurs dans la gestion des données
Global solution of Quadratic Problems by Interval Methods and Convex Relaxations
No full textInternational audienceInterval branch-and-bound solvers provide reliable algorithms for handling non-convex optimization problems by ensuring the feasibility and the optimality of the computed solutions, i.e. independently from the floating-point rounding errors. Moreover, these solvers deal with a wide variety of mathematical operators. However, these solvers are not dedicated to quadratic optimization and do not exploit nonlinear convex relaxations in their framework. We present an interval branch-andbound method that can efficiently solve quadratic optimization problems. At each node explored by the algorithm, our solver uses a quadratic convex relaxation which is as strong as a semi-definite programming relaxation, and a variable selection strategy dedicated to quadratic problems. The interval features can then propagate efficiently this information for contracting all variable domains. We also propose to make our algorithm rigorous by certifying firstly the convexity of the objective function of our relaxation, and secondly the validity of the lower bound calculated at each node. In the non-rigorous case, our experiments show significant speedups on general integer quadratic instances, and when reliability is required, our first results show that we are able to handle medium-sized instances in a reasonable running time
InchIGRAB: An Inchworm-Inspired Guided Retraction and Bending Device for Vine Robots During Colonoscopy
No full textInternational audienceVine robots are soft robots that translate by everting, or growing, from their tips. This mechanism of translation minimizes the application of shear forces on the environment, making them particularly well-suited for surgical tasks, such as colonoscopy. However, steering and retracting vine robots within tortuous and delicate environments presents significant challenges. In this paper, we introduce a novel soft robotic system for colonoscopy that consists of an inchworm-inspired device -the InchIGRAB -nested within a vine robot. The InchIGRAB is designed to enable steering of the vine robot along curved paths, as well as to enable controlled retraction of the vine robot after it reaches its target. We present the design, modeling, and fabrication of the robotic system and characterize its performance. Furthermore, we demonstrate the presented robotic system's ability to safely navigate the entire colon length, including several curved sections, during both forward motion and retraction, highlighting its potential as a robotic colonoscope that offers enhanced safety for patients
Nginx a 20 ans : un tour d'horizon
No full textNational audienceJe me souviens encore de la sortie de Nginx, juste après avoir passé plusieurs jours à configurer un serveur Apache. Je m’étais dit à cette époque (que les moins de 20 ans ne peuvent donc pas connaître) que malgré une consommation de ressources annoncée bien plus faible et la syntaxe plus lisible annoncée par ce nouvel outil, il était probablement moins fiable et que si Nginx tenait sur la durée, je lui donnerais sa chance… Il y a quelques années, j’ai donc tenu mon engagement. Aujourd’hui, il me paraît opportun de célébrer cet anniversaire dans les pages de notre magazine préféré
Accelerating Cell-Aware Model Generation for Sequential Cells using Graph Theory
No full textInternational audienceThe Cell-Aware (CA) methodology has become essential to detect and diagnose manufacturing intra-cell defects in modern semiconductor technologies. It characterizes standard cells by creating a defect-detection matrix, which serves as a reference that maps stimuli to the specific defects they can detect. Its limitation is that CA approach needs a number of time-consuming analog simulations to create the matrix. In [1] a graph-based methodology able to reduce the number of simulations to perform, called Transistor Undetectable Defect eLiminator (TrUnDeL), was presented. TrUnDeL can identify undetectable stimulus/defect pairs that are then excluded from the analog simulations. However, its use is limited to combinational cells and does not offer any guidance on handling sequential cells, which are usually the most complex cells. In this paper we present a new version of TrUnDeL that supports sequential cells analysis. Experiments conducted on sequential cells from two standard cell industrial libraries demonstrate that the CA generation time is reduced by 30% without compromising accuracy
Detection Of Generated Obscured Images Protecting Confidential Image Content
No full textInternational audienceMore and more multimedia data, such as images and videos, is transmitted over digital networks and stored or shared in the Cloud. For reasons of confidentiality or secret information, it is increasingly necessary to protect multimedia content directly. Although many image obscuration methods have been developed to protect the semantic content of images, few of them are both reversible and non-visible, and therefore detectable visually or by trained classifiers. In this paper, we propose a new image obscuration method based on variational autoencoders and a secret key, that transforms images from their source class into a target class, in a non-visible and reversible way, allowing the original image content to be recovered. In the experimental results we present whether the obscured images generated by our method are detectable