1,721,099 research outputs found
Modeling and Simulation of Bio-pathways using Hybrid Functional Petri Nets
Full text linkThe study of biological systems is growing rapidly, and can be considered as an intrinsic task in biological research and a prerequisite for diagnosing diseases and drug development. The integration of biological studies with computer technologies led to a noticeable development in this field with the appearance of many powerful modeling and simulation techniques and tools. The help of computers in biology resulted in deeper knowledge about complex biological systems and biopathways behaviors. Among modeling tools, the Petri Net formalism plays an important role. Petri Net is a powerful computerized and graphical modeling technique originally developed by Carl Adam Petri in 1960 to model discrete event systems. With its various extensions, Petri Nets find applications in many other fields including Biology. The extension known under the name Hybrid Functional Petri Net (HFPN) was developed specifically to model biological systems. Traditionally, biological processes are captured as systems of ordinary differential equations. However, HFPNs offer a much more elegant and versatile approach to represent these processes more accurately. In fact, these nets allow to capture phenomena which are impossible to capture with ordinary differential equations, while being more intuitive to understand and model with. In this work we propose an approach to automatically translate a system of ordinary differential equations representing a biological process into a HFPN. The resulting HFPN not only preserves the semantics of the original model, but is also more humanly readable thanks to the use of a novel technique to connect its components in a smart way. To validate our approach, we implemented it as an extension to the tool Real Time Studio (an integrated environment for modeling, simulation and automatic verification of real-time systems), and compared our simulation results with those obtained by simulating systems of ordinary differential equations on MATLAB
Préservation Numérique du Patrimoine Culturel : enrichissement et Reconstruction basés sur les CNNs Multimodaux Hiérarchiques et la Complétion d’Images
No full textLe patrimoine culturel joue un rôle important dans la définition de l'identité d’une société. La préservation physique à long terme du patrimoine culturel reste fragile et peut induire de multiples risques liés à la destruction et aux dommages accidentels. Les technologies numériques telles que la photographie et la numérisation 3D ont fourni de nouvelles alternatives pour la préservation numérique. Cependant, les adapter au contexte du patrimoine culturel est une tâche difficile. En effet, la numérisation complète des objets culturels (visuelle avec une copie digitale et historique avec des métadonnées) n'est facile que lorsqu'il s'agit d’objets physiquement en bon état possédant toutes leurs données (entièrement annotés). Cependant, dans le monde réel, de nombreux objets culturels souffrent de dégradation physique et de perte d'informations. Habituellement, pour annoter et conserver ces objets, les institutions culturelles font appel à des spécialistes de l'art, à des historiens et à d'autres institutions. Ce processus est fastidieux, nécessite beaucoup de temps et de ressources financières et peut souvent s’avérer inexact. Notre travail se concentre sur la préservation effective et rentable du patrimoine culturel, basée sur des méthodes avancées d'apprentissage automatique. L'objectif est de fournir un Framework à la phase d'enrichissement du processus de préservation numérique du patrimoine culturel. A travers cette thèse, nous proposons de nouvelles méthodes permettant d’améliorer le processus de préservation des objets culturels. Nos défis sont principalement liés au processus d'annotation et d'enrichissement des objets dont les données sont manquantes et/ou incomplètes (annotations et données visuelles) ; ce processus est souvent inefficace lorsqu’il est effectué manuellement. Nous introduisons diverses approches basées sur l'apprentissage automatique et l'apprentissage profond pour compléter automatiquement les données culturelles manquantes. Nous nous concentrons principalement sur deux types essentiels de données manquantes : les données textuelles (métadonnées) et les données visuelles.La première étape est principalement liée à l'annotation et à l'étiquetage des objets culturels à l'aide de l'apprentissage profond. Nous avons proposé des approches exploitant des caractéristiques visuelles et textuelles disponibles des objets culturels pour effectuer efficacement leur classification. (i) La première approche est proposée pour la Classification Hiérarchique des objets afin de mieux répondre aux exigences de métadonnées de chaque type d’objets et augmenter les performances de classification. (ii) La seconde approche est dédiée à la Classification Multimodale des objets culturels où un quelconque objet peut être représenté, lors de la classification, avec les métadonnées disponibles en plus de sa capture visuelle. La deuxième étape considère le manque d'informations visuelles lorsqu’il s’agit d’objets culturels incomplets et endommagés. Nous avons proposé dans ce cas, une approche basée sur l'apprentissage profond à travers des modèles génératifs et le clustering d’images pour effectuer la reconstruction visuelle d’objets culturels. Pour nos expérimentations, nous avons collecté une grande base de données culturelles mais nous avons sélectionné les tableaux d’arts pour nos tests et validations car ils possèdent une meilleure qualité d’annotation et sont donc mieux adapté pour mesurer les performances de nos algorithmes.Cultural heritage plays an important role in defining the identity of a society. Long-term physical preservation of cultural heritage remains risky and can lead to multiple problems related to destruction and accidental damage. Digital technologies such as photography and 3D scanning provided new alternatives for digital preservation. However, adapting them to the context of cultural heritage is a challenging task. In fact, fully digitizing cultural assets (visually and historically) is only easy when it comes to assets that are in a good physical shape and all their data is at possession (fully annotated). However, in the real-world, many assets suffer from physical degradation and information loss. Usually, to annotate and curate these assets, heritage institutions need the help of art specialists and historians. This process is tedious, involves considerable time and financial resources, and can often be inaccurate. Our work focuses on the cost-effective preservation of cultural heritage through advanced machine learning methods. The aim is to provide a technical framework for the enrichment phase of the cultural heritage digital preservation/curation process. Through this thesis, we propose new methods to improve the process of cultural heritage preservation. Our challenges are mainly related to the annotation and enrichment of cultural objects suffering from missing and incomplete data (annotations and visual data) which is often considered ineffective when performed manually. Thus, we propose approaches based on machine learning and deep learning to tackle these challenges. These approaches consist of the automatic completion of missing cultural data. We mainly focus on two types of missing data: textual data (metadata) and visual data.The first stage is mainly related to the annotation and labeling of cultural objects using deep learning. We have proposed approaches, that take advantage of cultural objects’ visual features as well as partially available textual annotations, to perform an effective classification. (i) the first approach is related to the Hierarchical Classification of Objects to better meet the metadata requirements of each cultural object type and increase the classification algorithm performance. (ii) the second proposed approach is dedicated to the Multimodal Classification of cultural objects where any object can be represented, during classification, with a subset of available metadata in addition to its visual capture. The second stage considers the lack of visual information when dealing with incomplete and damaged cultural objects. In this case, we proposed an approach based on deep learning through generative models and image data clustering to optimize the image completion process of damaged cultural heritage objects. For our experiments, we collected a large database of cultural objects. We chose to use fine-art paintings in our tests and validations as they were the best in terms of annotations quality
The Design for product service supportability (DfPSSu) methodology: Generating sector-specific guidelines and rules to improve product service systems (PSSs)
Full text linkNowadays manufacturers’ need to systematically develop innovative integrated solutions is increasingly pushed by new technologies, a multiple functionalities demand and a change in the customer value perception. For these reasons, it is very complex for Product Service Systems (PSS) providers to fulfil all the design requirements: designers must consider all the objectives the PSS wants to achieve during its whole lifecycle according to different criteria, which are often to be considered according to a trade-off balance. At present, Design for X (DfX) design methods represent the most important attempt to enhance product development according to certain characteristics or lifecycle phases: authors believe they can also support the PSS design, redesigning or enhancing products in certain X-dimensions, in particular those ones related to “service supportability”. On this basis, a methodology generating new Design for X (DfX) guidelines has been proposed: in this paper an application case in the mold industry shows how a physical product can be improved when a service has to be added and integrated. At the same time, new industry-specific PSS design guidelines and rules are proposed
A reference architecture for archival systems with application to product models
Full text linkPas de résumé en françaisNowadays, a major part of the information is in digital form. Digital preservation is essential to allowpeople to access information over time. From a computer science perspective, two major objectiveshave to be met to enable digital preservation: developing archival systems to manage the preserveddigital information, and select information representations that will facilitate the preservation. For complexinformation such as product models, these two objective are particularly hard to meet. Archivalsystems have to operate in a complex environment, interact with many different systems, and supportmay different business functions. Product model representations do not use all the possibilitiesof computer interpretation.Regarding the development of archival systems, the key is to determine what has to be described toprove that the archival system can effectively support the digital preservation. The Reference Modelfor an Open Archival Information System (OAIS) proposes a terminology to describe and comparearchives. The Audit and Certification of Trustworthy Digital Repository (ACTDR) provides criteria forthe certification of archives. One issue with these efforts is that there is not guidance on how to usethem within archival system descriptions.This thesis proposes a method called Reference Architecture for Archival Systems (RAAS) to describearchival systems implementations. RAAS relies on the DoD Architecture Framework to describethe various aspects of the archival systems. Moreover, RAAS provides an archival-specificterminology inspired by the OAIS Reference Model. RAAS also explains how the archival systemdescription can help for the ACTDR certification.RAAS is applied to a product model preservation case, to describe the various aspects of the archivalsystem. This description includes the interactions involving the archival systems, the archival systemfunctions, the definition of the preserved content, and the definition of the metadata. This descriptionformally refers to the OAIS terminology, and provides ACTDR certification evidence.This thesis also address the representation of product models by proposing the translation of productmodels from STEP to OWL. STEP is a standard for product model representation. The use ofOWL enables semantic relationship to enrich product information, and improve the search and theunderstanding of this information using data integration.The methodology used in this thesis can apply to other types of information, such as medical recordsDIJON-BU Doc.électronique (212319901) / SudocSudocFranceF
Système sémantique pour la customisation dynamique des modèles d'information de la PLM
No full textPas de résumé en françaisWe live in the information age. Data has become an essential asset for most everyday situations and business interactions. The need to share data, to generate information, and create new knowledge from that data is common to all fields of research and all economic activity. Managing data is a critical, and sometimes costly, process. When not properly defined, data might become incomplete, inconsistent or, even worse, unusable. Requirements for data evolve and we must define new data or update existing data over the entire data lifecycle. Evolving data requirements is an important issue and a technological challenge as it is not possible to define, in advance, information structures that meet requirements you do not yet know. Specifying information requirements is particularly challenging in domains such as manufacturing where information exchange involves many actors and sharing across multiple functions and software applications. As a result, it becomes hard to find a common information structure for representing data. The challenge is even bigger when a temporal aspect has to be considered since it requires the ability to extend the information structure dynamically over time. One area within the manufacturing domain that we have identified with these characteristics is Product Lifecycle Management (PLM). PLM involves many global actors using a myriad of software applications that perform a series of product management functions that can last from weeks to decades. Because the mechanism to extend models is static by its nature, requiring numerous updates of the initial information model, this operation is expensive in cost and time, and requires and understanding of the entire initial model to ensure correct extensions are developed. This research presents an alternative based on dynamic customization of information models in the context of PLM, by leveraging existing PLM standards and frameworks, and using emerging semantic web technologies such as OWL, SPARQL and SPIN. Following a state of the art in Chapter 2, Chapter 3 defines technical requirements used to evaluate existing PLM standards and frameworks. Based on the analysis of this evaluation, Chapter 4 presents new framework components for defining dynamically customizable information models for PLM. In chapter 5 these components are integrated together into a framework, and a use case demonstrates the efficiency of the framework. Chapter 6 concludes the research and introduces ideas for future research.DIJON-BU Doc.électronique (212319901) / SudocSudocFranceF
Interrogating witnesses for geometric constraint solving
No full textInternational audienceClassically, geometric constraint solvers use graph-based methods to decompose systems of geometric constraints. These methods have intrinsic limitations, which the witness method overcomes; a witness is a solution of a variant of the system. This paper details the computation of a basis of the vector space of free infinitesimal motions of a typical witness, and explains how to use this basis to interrogate the witness for dependence detection. The paper shows that the witness method detects all kinds of dependences: structural dependences already detectable by graph-based methods, but also non-structural dependences, due to known or unknown geometric theorems, which are undetectable by graph-based methods. It also discusses how to decide about the rigidity of a witness and how to decompose it
Moving object predictions in dynamic environments for autonomous ground vehicles
No full textLe rôle de la recherche dans le domaine des voitures autonomes est de pouvoir construire des systèmes physiques pouvant se déplacer dans un but précis, sans quelconque intervention extérieure humaine, que ce soit dans des environnements statiques et dynamiques, connus, partiellement connus et inconnus. Le domaine des véhicules autonomes continue à attirer chercheurs et autres techniciens. Le financement pour la recherche dans ce domaine a continué à prendre de l ampleur durant les dernières années, poussant ainsi les travaux à migrer de la théorie vers la pratique. Avant de pouvoir utiliser un véhicule autonome dans le monde réel, il est important d arriver à modéliser le véhicule dans une simulation d une part, et aussi de pouvoir évaluer les performances du véhicule dans le monde virtuel d autre part. Nous présentons le Framework PRIDE (Prediction In Dynamic Environments), une approche hiérarchique à multi résolutions pour la prédiction des objets mobiles. PRIDE regroupe plusieurs algorithmes de prédictions en une seule et même structure. PRIDE est basé sur 4D/RCS (Real-time Control System) et fournit des informations à des planificateurs à différents niveaux de granularité appropriés au temps de prédiction. Les plus bas niveaux du Framework utilisent des prédictions à courts termes basées sur le filtre étendu de Kalman associé à une mesure de confiance. Les plus hauts niveaux font appel à une approche de prédiction probabiliste basée sur la reconnaissance de situations associée à un model de coûts permettant de calculer des prédictions utilisant des informations et des contraintes associées à l environnement. PRIDE a connaissance de la structure du réseau routier via une base de données regroupant des informations sur l environnement. Le résultat de chaque prédiction est passé au planificateur pour le contrôle de la trajectoire du véhicule. Dans divers scénarios, nous avons utilisé PRIDE avec l outil de visualisation AutoSim dans un premier temps, et dans un deuxième temps, avec le Framework MOAST/USARSim, permettant ainsi à PRIDE de pouvoir considérer la physique, la cinématique ainsi que la dynamique des véhiculesThe goal of autonomous vehicles research is to build physical systems that can move purposefully and without human intervention in static and dynamic environments, and also in known, partially known and unknown environments. The field of autonomous vehicles is continuing to gain traction both with researchers and practitioners. Funding for research in this area has continued to grow over the past few years, and recent high profile funding opportunities have started to push theoretical research efforts into practical use. Before releasing any autonomous vehicle in the real world, it is important to model the components within a simulated environment and assess the performance of the vehicles in the virtual world. We present the PRIDE framework (Prediction In Dynamic Environments), a hierarchical multiresolutional approach for moving object prediction that incorporates multiple prediction algorithms into a single, unifying framework. PRIDE is based upon the 4D/RCS (Real-time Control System) and provides information to planners at the level of granularity that is appropriate for their planning horizon. The lower levels of the framework utilize estimation theoretic short-term predictions based upon an extended Kalman filter with an associated confidence measure. The upper levels utilize a probabilistic prédiction approach based upon situation recognition with an underlying cost model that provides predictions that incorporate environmental information and constraints. PRIDE is run in the systems' world model independently of the planner and the control system and has knowledge of the road structures via a road network database. The results of the prediction are made available to a planner to allow it to make accurate plans in dynamic environments. We have applied this approach to the visualization tool AutoSim and later on to the MOAST/USARSim framework which incorporates the physics, kinematics and dynamics of vehicles involved in traffic scenarios.DIJON-BU Sciences Economie (212312102) / SudocSudocFranceF
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
