Tesis sobre el tema "Réseau transformer"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 45 mejores tesis para su investigación sobre el tema "Réseau transformer".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
Cezar, Vinicius Oiring de Castro. "Contribution au renvoi de tension et à la reconstitution du réseau. Estimation des flux rémanents dans un transformateur". Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAT046/document.
Texto completoDuring the re-energization of the auxiliaries of a nuclear or hydraulic power plant, the most dangerous step is the re-energization of the power transformer, because of the temporary overvoltage and inrush currents. These transients phenomenon causes undesirable effects for both network and for the power transformer (electrodynamic forces over the windings, the magnetic circuit’s vibration, noise and the premature aging of the transformer). The goal of these thesis is to suggest new methodologies allowing us to evaluate unknown parameters (the residual flux’s values in the magnetic circuit before transformer’s energization). According to the latest problems in order to evaluate it (no direct method, derivation, voltage measurement error, etc) two new methods based on the previous magnetization of the magnetic circuit (prefluxing method) and on the leakage flux measurement of the magnetic circuit (direct measurement of the flux by measuring the magnetic induction method) are proposed
Cavallera, Didier. "Contribution au renvoi de tension et à la reconstitution du réseau. Identification des paramètres d'un réseau. Estimation des flux rémanents dans un transformateur". Phd thesis, Université de Grenoble, 2011. http://tel.archives-ouvertes.fr/tel-00732191.
Texto completoMessaoud, Kaouther. "Deep learning based trajectory prediction for autonomous vehicles". Electronic Thesis or Diss., Sorbonne université, 2021. http://www.theses.fr/2021SORUS048.
Texto completoThe trajectory prediction of neighboring agents of an autonomous vehicle is essential for autonomous driving in order to perform trajectory planning in an efficient manner. In this thesis, we tackle the problem of predicting the trajectory of a target vehicle in two different environments; a highway and an urban area (intersection, roundabout, etc.). To this end, we develop solutions based on deep machine learning by phasing the interactions between the target vehicle and the static and dynamic elements of the scene. In addition, in order to take into account the uncertainty of the future, we generate multiple plausible trajectories and the probability of occurrence of each. We also make sure that the predicted trajectories are realistic and conform to the structure of the scene. The solutions developed are evaluated using real driving datasets
Le, Moine Veillon Clément. "Neural Conversion of Social Attitudes in Speech Signals". Electronic Thesis or Diss., Sorbonne université, 2023. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2023SORUS034.pdf.
Texto completoAs social animals, humans communicate with each other by transmitting various types of information about the world and about themselves. At the heart of this process, the voice allows the transmission of linguistic messages denoting a strict meaning that can be decoded by the interlocutor. By conveying other information such as attitudes or emotions that connote the strict meaning, the voice enriches and enhances the communication process. In the last few decades, the digital world has become an important part of our lives. In many everyday situations, we are moving away from keyboards, mice and even touch screens to interactions with voice assistants or even virtual agents that enable human-like communication with machines. In the emergence of a hybrid world where physical and virtual reality coexist, it becomes crucial to enable machines to capture, interpret, and replicate the emotions and attitudes conveyed by the human voice.This research focuses on speech social attitudes, which can be defined - in a context of interaction - as speech dispositions towards others and aims to develop algorithms for their conversion. Fulfilling this objective requires data, i.e. a collection of audio recordings of utterances conveying various vocal attitudes. This research is thus built out of this initial step in gathering raw material - a dataset dedicated to speech social attitudes. Designing such algorithms involves a thorough understanding of what these attitudes are both in terms of production - how do individuals use their vocal apparatus to produce attitudes? - and perception - how do they decode those attitudes in speech? We therefore conducted two studies, a first uncovering the production strategies of speech attitudes and a second - based on a Best Worst Scaling (BWS) experiment - mainly hinting at biases involved in the perception such vocal attitudes, thus providing a twofold account for how speech attitudes are communicated by French individuals. These findings were the basis for the choice of speech signal representation as well as the architectural and optimisation choices for the design of a speech attitude conversion algorithm. In order to extend the knowledge on the perception of vocal attitudes gathered during this second study to the whole database, we worked on the elaboration of a BWS-Net allowing the detection of mis-communicated attitudes, and thus provided clean data for conversion learning. In order to learn how to convert vocal attitudes, we adopted a transformer-based approach in a many-to-many conversion paradigm with mel-spectrogram as speech signal representation. Since early experiments revealed a loss of intelligibility in the converted utterances, we proposed a linguistic conditioning of the conversion algorithm through incorporation of a speech-to-text module. Both objective and subjective measures have shown the resulting algorithm achieves better performance than the baseline transformer both in terms of intelligibility and attitude conveyed
Terrasse, Guillaume. "Géodétection des réseaux enterrés par imagerie radar". Thesis, Paris, ENST, 2017. http://www.theses.fr/2017ENST0015/document.
Texto completoThe thesis objective is to improve the different processing in order to make the data acquired by ground penetrating radar (B-scan) more understandable for the operators. Consequently, it will facilitate the pipe localisation. More particularly, we wish to highlight the hyperbolas in the B-scan because they point out the presence of a pipe. First of all, we are interested in removing all the useless information which might hide the hyperbolas. We proposed a filtering method removing unwanted reflections and noise. Then, we worked on an automatic hyperbola detection method and an estimation of their mathematical functions in quasi real time. Finally, we proposed a source separation method to distinguish the unwanted reflections from the hyperbolas with a minimal impact on them. This last work opens interesting perspectives in filtering, hyperbolas enhancement and hyperbola detection
Mohamed, Moussa Elmokhtar. "Conversion d’écriture hors-ligne en écriture en-ligne et réseaux de neurones profonds". Electronic Thesis or Diss., Nantes Université, 2024. http://www.theses.fr/2024NANU4001.
Texto completoThis thesis focuses on the conversion of static images of offline handwriting into temporal signals of online handwriting. Our goal is to extend neural networks beyond the scale of images of isolated letters and as well to generalize to other complex types of content. The thesis explores two distinct neural network-based approaches, the first approach is a fully convolutional multitask UNet-based network, inspired by the method of [ZYT18]. This approach demonstrated good results for skeletonization but suboptimal stroke extrac- tion. Partly due to the inherent temporal mod- eling limitations of CNN architecture. The second approach builds on the pre- vious skeletonization model to extract sub- strokes and proposes a sub-stroke level modeling with Transformers, consisting of a sub- stroke embedding transformer (SET) and a sub-stroke ordering transformer (SORT) to or- der the different sub-strokes as well as pen up predictions. This approach outperformed the state of the art on text lines and mathematical equations databases and addressed several limitations identified in the literature. These advancements have expanded the scope of offline-to-online conversion to include entire text lines and generalize to bidimensional content, such as mathematical equations
Terrasse, Guillaume. "Géodétection des réseaux enterrés par imagerie radar". Electronic Thesis or Diss., Paris, ENST, 2017. http://www.theses.fr/2017ENST0015.
Texto completoThe thesis objective is to improve the different processing in order to make the data acquired by ground penetrating radar (B-scan) more understandable for the operators. Consequently, it will facilitate the pipe localisation. More particularly, we wish to highlight the hyperbolas in the B-scan because they point out the presence of a pipe. First of all, we are interested in removing all the useless information which might hide the hyperbolas. We proposed a filtering method removing unwanted reflections and noise. Then, we worked on an automatic hyperbola detection method and an estimation of their mathematical functions in quasi real time. Finally, we proposed a source separation method to distinguish the unwanted reflections from the hyperbolas with a minimal impact on them. This last work opens interesting perspectives in filtering, hyperbolas enhancement and hyperbola detection
Fouineau, Alexis. "Méthodologies de Conception de Transformateurs Moyenne Fréquence pour application aux réseaux haute tension et réseaux ferroviaires". Thesis, Lyon, 2019. http://www.theses.fr/2019LYSE1256/document.
Texto completoMedium Frequency Transformers (MFT) are an innovative technology compared to low frequency transformers, with the promise of reduced volume and increased efficiency. This PhD thesis focuses in particular on their design for high voltage, high power applications, such as high voltage and medium voltage DC networks, as well as railway networks. In these applications, MFTs are used in converters that can generate specific constraints to be taken into account during their design: non-sinusoidal signals, polarization voltage, target inductance values. Moreover, the technological choices currently available for the realization of MFTs are numerous, and there is currently no consensus on any technology for any given application. Trends could be identified using a tool to classify MFT designs from the literature. Thus, the most promising technologies were selected and retained for the future. Based on these technologies, a design methodology was developed to quickly and semi-automatically design and compare MFTs with different technological choices. It consists of three steps: pre-design, analytical design, and validation. The complete analytical design of the MFT with different technological choices is carried out using an automated design tool developed during this thesis, named SUITED (SUpergrid Institute TransformEr Design). This methodology requires models and data for each of the components and phenomena of the MFT. Concerning the magnetic core, a review and selection of models from the literature were carried out for the evaluation of the magnetizing inductance and magnetic losses. In addition, magnetic characterizations have made it possible to highlight the impact of certain technological processes on the levels of loss of magnetic cores made of nanocrystalline material, which is an excellent candidate for MFTs. Concerning the windings, analytical models to calculate the magnetic field, leakage inductance and skin and proximity effects were developed and compared with those in the literature and simulations. These models are proving to be more accurate on the MFT geometries considered. On top of that, a new method for evaluating the parasitic capacitances of windings with rectangular turns has been successfully implemented and validated. Thermal networks have been identified for the different MFT geometries. The thermal resistances of conduction, convection and radiation are calculated from detailed models. In particular, the anisotropy of materials is taken into account for thermal conduction, and the convection coefficients are evaluated via different correlations for each face of the MFT. The thermal networks are then solved iteratively and analytically to take into account the non-linearity of the thermal resistances while optimizing the required computation time. Finally, this entire design methodology was applied to three case studies corresponding to the target applications: high voltage, medium voltage and rail. The results obtained do show the performance and necessity of this approach
Sibué, Jean-Romain. "Conception de systèmes d'alimentation sans contact pour la traction ferroviaire". Thesis, Grenoble, 2011. http://www.theses.fr/2011GRENT074/document.
Texto completoThe works presented in this thesis deal with the design and the sizing of magnetic component dedicated to contactless energy transfer for railway application like tramway. This family of component presents a strongly inductive behavior. A double resonance converter is used to supply and compensate reactive energy of transformer. In order to design this component and its associated converter, a design tool has been implemented. This one is based on analytical models of magnetic component and power electronic converter. One time designing realized, a study of losses, in windings and in magnetic cores, is realized by numerical simulation by using homogenization method (macroscopic representation of elements with electromagnetic complex properties). Then, the establishment of a model of thermal behavior of system is presented in order to guarantee desired working temperature. In order to check designing approach and tools, experimentations have been performed on prototypes of 1.6 and 100 kW. Obtained results show the accuracy and relevance of theoretical approach. Moreover, this study confirms the feasibility of this kind of high power system
Forthoffer, Martial. "Détection de contours par transformée en ondelettes et réseaux de neurones". Nancy 1, 1992. http://www.theses.fr/1992NAN10008.
Texto completoOussar, Yacine. "Réseaux d'ondelettes et réseaux de neurones pour la modélisation statique et dynamique de processus". Phd thesis, Université Pierre et Marie Curie - Paris VI, 1998. http://pastel.archives-ouvertes.fr/pastel-00000677.
Texto completoLefort, Romain. "Contribution des technologies CPL et sans fil à la supervision des réseaux de distribution d'électricité". Thesis, Poitiers, 2015. http://www.theses.fr/2015POIT2253/document.
Texto completoEstablishing a supervisory infrastructure allows a better smart management than an expensive strengthening of distribution network to respond to new constraints at the energies control (Consumption, REN, EV ...). To transmit data, Power Line Communication (PLC) technologies present an advantage in this context. In fact, it enables a superposition of High Frequency (HF) signals on electrical signal 50/60 Hz. However, electric networks have not been developed to this application because of difficult propagation conditions. This research work makes a contribution to develop a simulation platform in objective to transmit data to 1 MHz. In first time, each network element is studied singly and in second time, together, to estimate "Outdoor PLC" transmission performance. The first element studied is the networks variation in function of frequency and time. Several 24h disturbance measurements on LV customers are presented. The second element is the transformers which established connection between Medium Voltage (MV) and Low Voltage (LV). The proposed modeling method is based on a "lumped model" and a "black box model". These models are applied to a 100 kVA H61 transformer most commonly used by French distribution system operator in rural and suburban networks. The third element is the power line used in MV and LV networks. The proposed modeling method is based on a "cascaded model" from the theory of transmission line. This model is applied to one power line used in LV underground network. Each model is obtained from various impedance measurements. To complete, an introductory study on mobile radio communication is performed to remote network distribution
Barrère, Killian. "Architectures de Transformer légères pour la reconnaissance de textes manuscrits anciens". Electronic Thesis or Diss., Rennes, INSA, 2023. http://www.theses.fr/2023ISAR0017.
Texto completoTransformer architectures deliver low error rates but are challenging to train due to limited annotated data in handwritten text recognition. We propose lightweight Transformer architectures to adapt to the limited amounts of annotated handwritten text available. We introduce a fast Transformer architecture with an encoder, processing up to 60 pages per second. We also present architectures using a Transformer decoder to incorporate language modeling into character recognition. To effectively train our architectures, we offer algorithms for generating synthetic data adapted to the visual style of modern and historical documents. Finally, we propose strategies for learning with limited data and reducing prediction errors. Our architectures, combined with synthetic data and these strategies, achieve competitive error rates on lines of text from modern documents. For historical documents, they train effectively with minimal annotated data, surpassing state-ofthe- art approaches. Remarkably, just 500 annotated lines are sufficient for character error rates close to 5%
Makkaoui, Leila. "Compression d'images dans les réseaux de capteurs sans fil". Phd thesis, Université de Lorraine, 2012. http://tel.archives-ouvertes.fr/tel-00795503.
Texto completoBasmaji, Riad. "Etude des réseaux géodésiques réguliers : matrice de covariance des réseaux réguliers finis de type nivellement : application de la méthode de la transformée de Fourier à l'analyse : application de la méthode de la transformée de Fourier à l'analyse des réseaux réguliers infinis à une et deux variables". Observatoire de Paris, 1989. https://tel.archives-ouvertes.fr/tel-01958574.
Texto completoGliga, Lavinius ioan. "Diagnostic d'une Turbine Eolienne à Distance à l'aide du Réseau de Capteurs sans Fil". Thesis, Normandie, 2019. http://www.theses.fr/2019NORMR063/document.
Texto completoDirect Drive Wind Turbines (DDWTs) are equipped with Permanent Magnet Syn- chronous Generators (PMSGs). Their three most common failures are demagnetization, ec- centricity (static, dynamic and mixed) and inter-turn short circuit. Machine Current Signa- ture Analysis is often used to look for generator problems, as these impairments introduce additional harmonics into the generated currents. The Fast Fourier Transform (FFT) is utilized to compute the spectrum of the currents. However, the FFT calculates the whole spectrum, while the number of possible faults and the number of introduced harmonics is low. The Goertzel algorithm, implemented as a filter (the Goertzel filter), is presented as a more efficient alternative to the FFT. The spectrum of the currents changes with the wind speed, and thus the detection is made more difficult. The Extended Kalman Filter (EKF) is proposed as a solution. The spectrum of the residuals, computed between the estimated and the generated current, is constant, regardless of the wind speed. However, the effect of the faults is visible in the spectrum. When using the EKF, one challenge is to find out the covariance matrix of the process noise. A new method was developed in this regard, which does not use any of the matrices of the filter. DDWTs are either placed in remote areas or in cities. For the monitoring of a DDWT, tens or hundreds of kilometers of cables are necessary. Wireless Sensor Networks (WSNs) are suited to be used in the communication infrastructure of DDWTs. WSNs have lower initial and maintenance costs, and they are quickly installed. Moreover, they can complement wired networks. Different wireless technologies are com- pared - both wide area ones, as well as short range technologies which support high data rates
Dugas, Alexandre. "Architecture de transformée de cosinus discrète sur deux dimensions sans multiplication et mémoire de transposition". Mémoire, Université de Sherbrooke, 2012. http://hdl.handle.net/11143/6174.
Texto completoSaadaoui, Safa. "Conception d'un système de communication sans fil industriel basé sur la transformée en ondelettes". Thesis, Université de Lorraine, 2019. http://www.theses.fr/2019LORR0048/document.
Texto completoIn this thesis, we will present a multi-user communication architecture based on wireless sensor networks in a noisy industrial environment. Two modes of operation of this architecture are presented ; a Many-To-One mode linking several transmitter sensors to a single receiver and a One-To-Many mode linking a transmitter to several receiver sensors. The physical layer of this system is based on the inverse transform (IDWPT) at transmission and the discrete wavelet packet transform (DWPT) at reception. To test our architecture, an industrial channel model is proposed that takes into account the phenomena of multipath and fading in addition to additive noise. The latter being modelled as Gaussian noise to which is added an impulse noise causing significant signal degradation. The architecture is tested for different wireless communication configurations and wavelet shapes to provide an optimal communication mode. Also, an improvement in the robustness of our system is achieved by adding channel error correction coding and pulse noise thresholding at reception to minimize the effects of industrial noise on the received signals. Using an error-correcting code, the detection and reconstruction of signals is error-free from an SNR of 8dB for a coding rate of 1/4 for transmission on a fading channel. For an optimal receiver based on noise thresholding, the performance in terms of binary error rates is improved by 10dB for transmission over an industrial noise channel. Finally, a comparison of the robustness of our pulse architecture with a system based on a conventional OFDM multi-carrier modulation is carried out. This leads us to propose a robust multi-user communication system based on wireless sensor networks for communications in difficult industrial environments
Madre, Guillaume. "Application de la transformée en nombres entiers à l'étude et au développement d'un codeur de parole pour transmission sur réseaux IP". Brest, 2004. http://www.theses.fr/2004BRES2036.
Texto completoOur study considers the vocal signals compression for the transmission of Voice over Internet Protocol (VoIP). The prospects being the implementation of a telephony IP application, the work provides the first elements for a real-time speech coding system and its integration to a DSP. They are concentrated on the speech CS-ACELP (Conjugate Structure- Algebraic Code-Excited Linear Prediction) G. 729 coder, retained among the International Telecommunications Union (ITU) recommendations and already recognized for its low implementation complexity. The main aspect was to improve its performances and to decrease its computational cost, while maintaining the compromise between the coding quality and the required complexity. To reduce the computational cost of this coder, we looked further into the mathematical bases of the Number Theoretic Transform (NTT) which is brought to find more and more various applications in signal processing. We introduced more particularly the Fermat Number Transform (FNT) which is well suited for digital processing operations. Its application to different coding algorithms allows an important reduction of the computational complexity. Thus, the development of new efficient algorithms, for the Linear Prediction (LP) of the speech signal and the excitation modeling, has allowed a modification of the G. 729 coder and his implementation on a fixed-point processor. Moreover, a new function of Voice Activity Detection (VAD) has carried out the implementation of one more efficient procedure for silences compression and the reduction of the transmission rate
Dubois, Patrick. "Real-time FPGA Radiometric Calibration of an Imaging Fourier Transform Spectrometer". Thesis, Université Laval, 2009. http://www.theses.ulaval.ca/2009/26685/26685.pdf.
Texto completoDouillard, Arthur. "Continual Learning for Computer Vision". Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS165.
Texto completoI first review the existing methods based on regularization for continual learning. While regularizing a model's probabilities is very efficient to reduce forgetting in large-scale datasets, there are few works considering constraints on intermediate features. I cover in this chapter two contributions aiming to regularize directly the latent space of ConvNet. The first one, PODNet, aims to reduce the drift of spatial statistics between the old and new model, which in effect reduces drastically forgetting of old classes while enabling efficient learning of new classes. I show in a second part a complementary method where we avoid pre-emptively forgetting by allocating locations in the latent space for yet unseen future class. Then, I describe a recent application of CIL to semantic segmentation. I show that the very nature of CSS offer new specific challenges, namely forgetting on large images and a background shift. We tackle the first problem by extending our distillation loss introduced in the previous chapter to multi-scales. The second problem is solved by an efficient pseudo-labeling strategy. Finally, we consider the common rehearsal learning, but applied this time to CSS. I show that it cannot be used naively because of memory complexity and design a light-weight rehearsal that is even more efficient. Finally, I consider a completely different approach to continual learning: dynamic networks where the parameters are extended during training to adapt to new tasks. Previous works on this domain are hard to train and often suffer from parameter count explosion. For the first time in continual computer vision, we propose to use the Transformer architecture: the model dimension mostly fixed and shared across tasks, except for an expansion of learned task tokens. With an encoder/decoder strategy where the decoder forward is specialized by a task token, we show state-of-the-art robustness to forgetting while our memory and computational complexities barely grow
Ouillon, Guy. "Application de l'analyse multifractale et de la transformée en ondelettes anisotropes à la caractérisation géométrique multi-échelle des réseaux de failles et de fractures". Nice, 1995. http://www.theses.fr/1995NICE4834.
Texto completoKamienny, Pierre-Alexandre. "Efficient adaptation of reinforcement learning agents : from model-free exploration to symbolic world models". Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS412.
Texto completoReinforcement Learning (RL) encompasses a range of techniques employed to train autonomous agents to interact with environments with the purpose of maximizing their returns across various training tasks. To ensure successful deployment of RL agents in real-world scenarios, achieving generalization and adaptation to unfamiliar situations is crucial. Although neural networks have shown promise in facilitating in-domain generalization by enabling agents to interpolate desired behaviors, their limitations in generalizing beyond the training distribution often lead to suboptimal performance on out-of-distribution data. These challenges are further amplified in RL settings characterized by non-stationary environments and constant distribution shifts during deployment. This thesis presents novel strategies within the framework of Meta-Reinforcement Learning, aiming to equip RL agents with the ability to adapt at test-time to out-of-domain tasks. The first part of the thesis focuses on model-free techniques to learn effective exploration strategies. We consider two scenarios: one where the agent is provided with a set of training tasks, enabling it to explicitly model and learn generalizable task representations; and another where the agent learns without rewards to maximize its state coverage. In the second part, we investigate into the application of symbolic regression, a powerful tool for developing predictive models that offer interpretability and exhibit enhanced robustness against distribution shifts. These models are subsequently integrated within model-based RL agents to improve their performance. Furthermore, this research contributes to the field of symbolic regression by introducing a collection of techniques that leverage Transformer models, enhancing their accuracy and effectiveness. In summary, by addressing the challenges of adaptation and generalization in RL, this thesis focuses on the understanding and application of Meta-Reinforcement Learning strategies. It provides insights and techniques for enabling RL agents to adapt seamlessly to out-of-domain tasks, ultimately facilitating their successful deployment in real-world scenarios
El, alaoui Imane. "Transformer les big social data en prévisions - méthodes et technologies : Application à l'analyse de sentiments". Thesis, Angers, 2018. http://www.theses.fr/2018ANGE0011/document.
Texto completoExtracting public opinion by analyzing Big Social data has grown substantially due to its interactive nature, in real time. In fact, our actions on social media generate digital traces that are closely related to our personal lives and can be used to accompany major events by analysing peoples' behavior. It is in this context that we are particularly interested in Big Data analysis methods. The volume of these daily-generated traces increases exponentially creating massive loads of information, known as big data. Such important volume of information cannot be stored nor dealt with using the conventional tools, and so new tools have emerged to help us cope with the big data challenges. For this, the aim of the first part of this manuscript is to go through the pros and cons of these tools, compare their respective performances and highlight some of its interrelated applications such as health, marketing and politics. Also, we introduce the general context of big data, Hadoop and its different distributions. We provide a comprehensive overview of big data tools and their related applications.The main contribution of this PHD thesis is to propose a generic analysis approach to automatically detect trends on given topics from big social data. Indeed, given a very small set of manually annotated hashtags, the proposed approach transfers information from hashtags known sentiments (positive or negative) to individual words. The resulting lexical resource is a large-scale lexicon of polarity whose efficiency is measured against different tasks of sentiment analysis. The comparison of our method with different paradigms in literature confirms the impact of our method to design accurate sentiment analysis systems. Indeed, our model reaches an overall accuracy of 90.21%, significantly exceeding the current models on social sentiment analysis
Kizil, Hüseyin Christophe. "Conception d'une architecture numérique configurable pour la transformée en paquets d'ondelettes discrètes : application aux transmissions numériques ultra-large bande impulsionnelles". Electronic Thesis or Diss., Université de Lorraine, 2019. http://www.theses.fr/2019LORR0265.
Texto completoImpulse radio ultra-wideband communications (IR-UWB) are data transmissions in 3.6 - 10.1 GHz frequency band for application fields of very short-range wireless communications, body area networks (BAN) or high precision location. The advantage of IR-UWB is the possibility of getting a high throughput and the signals have immunity to noise and multipath. Nevertheless, the main drawback is that kind of communication requires complex and energy consuming circuits. In this research works, we propose an architectures based on wavelet packets based on the inverse discrete wavelet packet transform (IDWPT) performing as a pulse generator reducing the complexity of architectures while preserving proper performance. Similarly, the direct transform (DWPT) can be used as decoder of the pulses. We have studied the feasibility of such architectures and develop a sensor network for integration into a BAN. The architecture of the pulse generator and the decoder are based on banks of interpolation and decimation filters. These filter banks have a tree structure which is symmetrical and regular providing a flexibility. The first tests of transmissions through a Gaussian channel under MATLAB - Simulink environment for different wavelets (Haar, Db2, Db4, Coif2 and rbio1.5) show that it is possible to consider such communication based on wavelet packets. Through a Gaussian channel, all the studied wavelets have the same performances. Therefore, it is preferable to choose the smallest architecture, that based on the Haar wavelet. Since the architectures are based exclusively on interpolation and decimation filters suitable for FPGA integration considering transposed filters with a polyphase structure. The detailed implementation of wavelet-based emitters and receivers studied is presented in this manuscript and compared to the previous works. We were able to observe wavelet packets from experiments that were conducted to wide bande frequencies. In this work, the time-shift problem is also considered and resolved by proposing an specific algorithm based on the uniqueness of received packets by the receiver and corresponding to a linear combination of packets coming from transmitters. In addition, a mathematical formalization validates the algorithm and extends the method for all orthogonal and compact wavelets having the same characteristics as Daubechies wavelets
Grall, Céline. "La Faille Nord Anatolienne dans sa portion immergée en mer de Marmara : évolution du réseau de failles et migration de fluides". Thesis, Aix-Marseille, 2013. http://www.theses.fr/2013AIXM4371.
Texto completoThis study addresses the issue on the deformation and the fluid migration, associated to the North Anatolian Fault within the Sea of Marmara (Turkey).First, we aim to constrain the evolution of the fault network and the slip rate through time, by two independent approaches: - historical thermal modeling of a basin of the Sea of Marmara; - definition of a Mass Transport Deposit as a fault lateral slip marker, and dated by stratigraphic interpretation. We show that: - (1) the present day fault system, formed by a main fault which accommodated the main part of the inter-plate deformation does not significantly evolved since 330.000 ± 100.000 years - (2) a progressive reorganization of the fault network occurred since the last 2.5-1.5 Ma.Secondly, we discuss the triggers of Mass Transport Processes. We show that: - (1) despite submarine mass movements are related to tectonic activity (mainly earthquakes and crustal stretching), their frequency and their size are also modulated by glacio-eustatic changes; -(2) remarkable Mass Transport Deposits display some cyclicity in stratigraphic sequences which are apparently correlated to transitions between salty marine and lacustrine environments. This cyclicity is perhaps explained by marine clay activity (swelling) under low brackish-fresh water conditions, which can trigger sediment destabilization.Third, we investigate the diversity of active fluid seepages contexts. We propose that the widespread occurrence of fluid expulsion sites can be explained by up-dip gas migration by buoyancy along permeable strata toward their edges, and along fractures within the basement toward both the edges of the basins and topographic highs
Peteri, Renaud. "Extraction de réseaux de rues en milieu urbain à partir d'images satellites à très haute résolution spatiale". Phd thesis, École Nationale Supérieure des Mines de Paris, 2003. http://pastel.archives-ouvertes.fr/pastel-00000508.
Texto completoKhalil, Aamre. "Architectures fonctionnelles de transmission sans fils adaptées aux milieux industriels fortement bruités". Electronic Thesis or Diss., Université de Lorraine, 2020. http://www.theses.fr/2020LORR0183.
Texto completoRecent years have been marked by rapid and profound changes in data transmission technologies. These changes are dictated by the massive increase in data rates, the diversity of physical media, the multiplication of joint transmission standards and interoperability requirements, as well as the increase in constraints linked to the quality of service, which are reflected in particular in operational safety constraints (reliability, fault tolerance). In the field of wireless communications, these aspects are compounded by the use of an already congested and non-extensible frequency spectrum, as well as the need to take account of the mobility, autonomy and safety of users. Inevitably, all this leads to very strong constraints in the design of transmission circuits and systems that must support the new transmission protocols. This is particularly the case for all aspects relating to error-correcting codes responsible for protecting communications against external disturbances likely to alter the information transmitted. In fact, the processing algorithms to be implemented, which are very demanding in terms of throughput and complexity, make the existing coding/decoding architectures unsuitable because of their insufficient computing capacity. The aim of the thesis is in the continuation of previous work that has been the subject of several theses and publications concerning architectures dedicated to the processing of error-correcting codes, processor architectures and on-chip network architectures, while focusing on the fault tolerance aspects. The overall objective here is to design new functional digital architectures based on discrete wavelet packet modulation, using Inverse Discrete Wavelet Packet Transform (IDWPT) in transmission, and Discrete Wavelet Packet Transform (DWPT) in reception, exploiting two transmission modes: One-to-Many (OtM) for transmission between a single transmitter and several sensors ranging from 8, 16 up to 32 or more, and a Many-to-One (MtO) mode for transmission between several users and a single receiver with the same possible configuration, followed by the integration of modern correction codes in our architecture (notably the different families of Reed Solomon codes and Convolutional Codes), offering high processing power and versatility (flexibility and configurability), for high noise industrial environments. The use of error-correcting codes allowed us to have a gain of 4 to 6dB for Reed Solomon codings and convolutional codes used individually, 5 to 10db with the combination of these two codes and 2 to 4dB more for shortened Reed Solomon codes
Piat, Guilhem Xavier. "Incorporating expert knowledge in deep neural networks for domain adaptation in natural language processing". Electronic Thesis or Diss., université Paris-Saclay, 2023. http://www.theses.fr/2023UPASG087.
Texto completoCurrent state-of-the-art Language Models (LMs) are able to converse, summarize, translate, solve novel problems, reason, and use abstract concepts at a near-human level. However, to achieve such abilities, and in particular to acquire ``common sense'' and domain-specific knowledge, they require vast amounts of text, which are not available in all languages or domains. Additionally, their computational requirements are out of reach for most organizations, limiting their potential for specificity and their applicability in the context of sensitive data.Knowledge Graphs (KGs) are sources of structured knowledge which associate linguistic concepts through semantic relations. These graphs are sources of high quality knowledge which pre-exist in a variety of otherwise low-resource domains, and are denser in information than typical text. By allowing LMs to leverage these information structures, we could remove the burden of memorizing facts from LMs, reducing the amount of text and computation required to train them and allowing us to update their knowledge with little to no additional training by updating the KGs, therefore broadening their scope of applicability and making them more democratizable.Various approaches have succeeded in improving Transformer-based LMs using KGs. However, most of them unrealistically assume the problem of Entity Linking (EL), i.e. determining which KG concepts are present in the text, is solved upstream. This thesis covers the limitations of handling EL as an upstream task. It goes on to examine the possibility of learning EL jointly with language modeling, and finds that while this is a viable strategy, it does little to decrease the LM's reliance on in-domain text. Lastly, this thesis covers the strategy of using KGs to generate text in order to leverage LMs' linguistic abilities and finds that even naïve implementations of this approach can result in measurable improvements on in-domain language processing
Chaniot, Johan. "Caractérisation morphologique efficace de matériaux par cartes de distance". Thesis, Lyon, 2019. http://www.theses.fr/2019LYSES039.
Texto completoIn a technologically advanced world, energy consumption is rapidly increasing deepening the ongoing environmental crisis. Therefore, solutions must be found to provide the required energy, while reducing greenhouse gas emissions.Catalysis is an excellent way to improve the energy efficiency of industrial processes. Heterogeneous catalysts, here porous microstructures, are at the heart of this process, particularly for refining and petrochemical industry, specifically for biofuel generation.Their morphological description provides key information. Thus far, correlations have been established between the structural properties and performance of these materials. Nevertheless, heterogeneous catalysis is a very complex process and the traditional numerical descriptors provide insufficient information and fail to assist in material selection. The work addressed in this thesis aims to develop new digital descriptors of microstructures that are easily interpretable, efficient and complementary to the current state-of-the-art solutions. The objective is to complete the set of descriptors, to help in the optimal selection of the appropriate catalysts for a given application. More specifically, we focus our work on a geometric and topological characterization of the porous network, without taking into account physicochemical phenomena, and not being limited by the complexity of the microstructure studied.Our different approaches focus on the concepts of percolation, ability to cross a microstructure; geometric tortuosity, sinuosity and interconnectivity of a network; and heterogeneity. The geometric and topological characteristics linked to the Minkowski functionals in 3D are fitted to catalysis field by estimating accessibility of a microstructure for a given sphere size (A-protocol), described by morphological erosion efficiently calculated by distance maps. To characterize the pore topology, we define an operator, the M-tortuosity, that can be applied to any segmented volume, without arbitrarily defining source points or planes. We propose an efficient M-tortuosity estimator by calculating distance maps; which is then generalized by power factors. This operator is then extended to distinct ways.First, to the case of a probe of finite size (M-tortuosity-by-iterative-erosions), characterizing bottleneck effects which are usually quantified using constrictivity. Then, to characterize the spatial scale dependence of tortuosity (H-tortuosity), characterizing, among others, the heterogeneity of the structure. Finally, both aspects are gathered into the H-tortuosity-by-iterative-erosions.Secondly, these ensemble operators, suitable for binary images, are extended to the functional case, to discriminate grayscale images (F-tortuosity and HF-tortuosity). These functional extensions have various purposes: combining local information with tortuosity assessment of the overall structure, and characterizing tomographic images without accurate segmentation.The discrimination power of these operators, ensemble and functional, is assessed on toy cases and on multi-scale Boolean Cox schemes. Moreover, their similarities and complementarities are analysed using these very same stochastic models.In catalysis and biocatalysis, three types of catalysts are considered: zeolites, MOFs (Metal-Organic Framework) and aluminas. These applications highlight their wide scope, and lead to consider their usefulness out of catalyst domain; in neuroscience and for turbid media characterization
Lancry, Matthieu. "Contribution à l'étude de méthodes de sensibilisation de fibres optiques de communication et mécanismes associés". Lille 1, 2004. https://ori-nuxeo.univ-lille1.fr/nuxeo/site/esupversions/ede850ab-f1c9-467d-9f5c-9ca2b64ac672.
Texto completoWang, Kun. "Algorithmes et méthodes pour le diagnostic ex-situ et in-situ de systèmes piles à combustible haute température de type oxyde solide". Phd thesis, Université de Franche-Comté, 2012. http://tel.archives-ouvertes.fr/tel-01017170.
Texto completoMammeri, Abdelhamid. "Compression et transmission d'images avec énergie minimale application aux capteurs sans fil". Thèse, Université de Sherbrooke, 2010. http://hdl.handle.net/11143/5800.
Texto completoTerrissa, Saddek Labib. "Vision bio-fondée appliquée à l'analyse d'images". Le Havre, 2006. http://www.theses.fr/2006LEHA0001.
Texto completoThe environment recognition goes inevitably through the retina of alive beings. This component is a part of the central nervous system witch performs many processing, especially in motion detection and features extraction of a scene. These aspects are not reach the development level compared with static processing in artificial vision. For these reasons, we are developed a generic prototype of artificial retina inspired from biology witch we are titled Retina. This system is based on artificial neural network organised to simulate the radial variation of the acuity. The information is encoded through the implementation of analogue and impulse neurons. These retinal signal impulses are analysed in order to detect the variation of grey level in dynamic images. Moreover, we reconstitute these images using neural networks techniques and signal processing
Rekik, Siwar. "Sécurisation de la communication parlée par une techhnique stéganographique". Thesis, Brest, 2012. http://www.theses.fr/2012BRES0061.
Texto completoOne of the concerns in the field of secure communication is the concept of information security. Today’s reality is still showing that communication between two parties over long distances has always been subject to interception. Providing secure communication has driven researchers to develop several cryptography schemes. Cryptography methods achieve security in order to make the information unintelligible to guarantee exclusive access for authenticated recipients. Cryptography consists of making the signal look garbled to unauthorized people. Thus, cryptography indicates the existence of a cryptographic communication in progress, which makes eavesdroppers suspect the existence of valuable data. They are thus incited to intercept the transmitted message and to attempt to decipher the secret information. This may be seen as weakness in cryptography schemes. In contrast to cryptography, steganography allows secret communication by camouflaging the secret signal in another signal (named the cover signal), to avoid suspicion. This quality motivated the researchers to work on this burning field to develop schemes ensuring better resistance to hostile attackers. The word steganography is derived from two Greek words: Stego (means cover) and graphy (means writing). The two combined words constitute steganography, which means covert writing, is the art of hiding written communications. Several steganography techniques were used to send message secretly during wars through the territories of enemies. The major contributions of this thesis are the following ones. We propose a new method to secure speech communication using the Discrete Wavelet Transforms (DWT) and the Fast Fourier Transform (FFT). Our method exploits first the high frequencies using a DWT, then exploits the low-pass spectral properties of the speech magnitude spectrum to hide another speech signal in the low-amplitude high-frequencies region of the cover speech signal. The proposed method allows hiding a large amount of secret information while rendering the steganalysis more complex. Comparative evaluation based on objective and subjective criteria is introduced for original speech signal, stego-signal and reconstructed secret speech signal after the hiding process. Experimental simulations on both female and male speakers revealed that our approach is capable of producing a stego speech that is indistinguishable from the cover speech. The receiver is still able to recover an intelligible copy of the secret speech message. We used an LPC10 coder to test the effect of the coding techniques on the stego-speech signals. Experimental results prove the efficiency of the used coding technique since intelligibility of the stego-speech is maintained after the encoding and decoding processes. We also advocate a new steganalysis technique to ensure the robustness of our steganography method. The proposed classifier is called Autoregressive time delay neural network (ARTDNN). The purpose of this steganalysis system is to identify the presence or not of embedded information, and does not actually attempt to extract or decode the hidden data. The low detecting rate prove the robustness of our hiding technique
Ghosn, Ali. "Un nouveau calcul pour la forme d'intersection d'une variété projective". Nancy 1, 1988. http://www.theses.fr/1988NAN10137.
Texto completoDe, Brito Milena. "Investigation of interpenetrating polymer networks and recent UV curable chemistries". Thesis, Mulhouse, 2011. http://www.theses.fr/2011MULH5772/document.
Texto completoThe aim of this thesis is to develop and characterize UV curable resins for rapid prototyping application. The study started with the optimization of commercially available hybrid acrylate/epoxide systems (IPNs) commonly used in this industrial field. The reactivity of the different tested formulations in conjunction with their thermomechanical properties have been assessed by means of Real time Fourier transform infrared spectroscopy (RT FTIR) and Dynamic mechanical analysis (DMA). In the meantime, a more academic study has been performed on an epoxide/methacrylate mixture in order to get a better understanding and a control of the IPN formation. The influence of the light intensity and the photoinitiator concentration on the final properties has been examined. Then, UV curable resins whose chemistry is less conventional have been considered. Owing to the lack of commercial availability of some monomers, time has been spent to synthesize them. Some formulations display promising features especially in terms of reactivity, thermomechanical properties and linear shrinkage. Finally, a simple method to measure linear shrinkage has been set up to compare the different systems proposed during the thesis and thus evidences the less shrinkable UV curable system
Elouali, Fatima-Zohra. "Élaboration et caractérisation de gels de polymères acryliques et méthacryliques en présence de solvants isotropes et anisotropes". Thesis, Lille 1, 2009. http://www.theses.fr/2009LIL10175.
Texto completoThis study focuses on the development and characterization of systems based on biocompatible polymers in the presence of isotropic and anisotropic solvents. The characterization of these networks of acrylic and methacrylic polymers obtained by radical photopolymerization was carried out by swelling experiments in an isotropic environment (water) and in anisotropic solvents (using cyanobiphenyls as liquid crystals). The kinetics of swelling in water was followed by the rate of crosslinking and pH. The swelling of polymer networks in some liquid crystals has been studied as a function of temperature using an optical microscope and a heating stage. The swelling ratio did not exceed 5% because of strong interactions between the polymer and liquid crystal.These materials are of particular interest in the field of electro-optical devices. A detailed study of electro-optical properties has been performed for polymer-liquid crystal films, prepared by the method of phase separation induced by ultraviolet (UV) radiation. Some binary systems exhibit relaxation phenomena leading to optical memory effects after stopping the electric field. The characterization of synthesized structures was performed through the use of powerful experimental techniques such as Fourier Transform Infrared (FTIR) spectroscopy, Nuclear Magnetic Resonance (NMR), Polarized Optical Microscopy (MOP), Scanning Electron Microscopy (SEM), Size Exclusion Chromatography (SEC) and Differential Scanning Calorimetry (DSC)
Morabit, Safaa El. "New Artificial Intelligence techniques for Computer vision based medical diagnosis". Electronic Thesis or Diss., Valenciennes, Université Polytechnique Hauts-de-France, 2023. http://www.theses.fr/2023UPHF0013.
Texto completoThe ability to feel pain is crucial for life, since it serves as an early warning system forpotential harm to the body. The majority of pain evaluations rely on patient reports. Patients who are unable to express their own pain must instead rely on third-party reportsof their suffering. Due to potential observer bias, pain reports may contain inaccuracies. In addition, it would be impossible for people to keep watch around the clock. Inorder to better manage pain, especially in noncommunicative patients, automatic paindetection technologies might be implemented to aid human caregivers and complementtheir service. Facial expressions are used by all observer-based pain assessment systemsbecause they are a reliable indicator of pain and can be interpreted from a distance.Taking into consideration that pain generally generates spontaneous facial behavior,these facial expressions could be used to detect the presence of pain. In this thesis, weanalyze facial expressions of pain in order to address pain estimation. First, we presenta thorough analysis of the problem by comparing numerous common CNN (Convolutional Neural Network) architectures, such as MobileNet, GoogleNet, ResNeXt-50, ResNet18, and DenseNet-161. We employ these networks in two unique modes: standalone and feature extraction. In standalone mode, models (i.e., networks) are utilized to directly estimate pain. In feature extractor mode, "values" from the middle layer are extracted and fed into classifiers like Support Vector Regression (SVR) and Random Forest Regression (RFR).CNNs have achieved significant results in image classification and have achievedgreat success. The effectiveness of Transformers in computer vision has been demonstrated through recent studies. Transformer-based architectures were proposed in the second section of this thesis. Two distinct Transformer-based frameworks were presented to address two distinct pain issues: pain detection (pain vs no pain) and thedistinction between genuine and posed pain. The innovative architecture for binaryidentification of facial pain is based on data-efficient image transformers (Deit). Twodatasets, UNBC-McMaster shoulder pain and BioVid heat pain, were used to fine-tuneand assess the trained model. The suggested architecture is built on Vision Transformers for the detection of genuine and simulated pain from facial expressions (ViT). Todistinguish between Genuine and Posed Pain, the model must pay particular attentionto the subtle changes in facial expressions over time. The employed approach takes intoaccount the sequential aspect and captures the variations in facial expressions. Experiments on the publicly accessible BioVid Heat Pain Database demonstrate the efficacy of our strategy
Benaissa, Djamila. "Approche théorique et expérimentale du comportement électro-optique des systèmes polymères/cristaux liquides". Thesis, Lille 1, 2009. http://www.theses.fr/2009LIL10126.
Texto completoA study of materials based on polymers and liquid crystals of type PDLC (for Polymer Dispersed Liquid Crystals), elaborated by the method of phase separation induced by ultraviolet radiation (UV), was carried out for mixtures containing the nematic liquid crystal E7 and the monomer tripropyleneglycoldiacrylate (TPGDA). These materials possess interesting electro-optical features in particular for privacy windows with controlled transparency.An analysis by infrared spectroscopy of the elaborated polymer networks allowed to deduce that a quasi-total conversion of the reactive acrylic functions of the monomer was obtained for a mixture containing 70 % of liquid crystal and 30 % of monomer. These materials were, then, studied by various techniques of characterization, such as differential scanning calorimetry, polarized optical microscopy, scanning electron microscopy, and UV-visible spectroscopy, who allowed to obtain informations on the thermophysical, morphological, and spectroscopical properties of these systems.An attempt was made to rationalize the electro-optical response of PDLC films by using a simple model, based on a hierarchy of order parameters. Using the theoretical description from two theoretical approximations (RGA and ADA), this model gave a proper description of the electro-optical behaviour of these complex systems. The study led on the confrontation of this model to experimental results, allowing to obtain some interesting results which are useful in understanding and improvement of the electro-optical response of PDLC films
Calvin, Christophe. "Minimisation du sur-coût des communications dans la parallélisation des algorithmes numériques". Phd thesis, Grenoble INPG, 1995. http://tel.archives-ouvertes.fr/tel-00005034.
Texto completoMartin, Stéphane. "Edition collaborative des documents semi-structurés". Phd thesis, Université de Provence - Aix-Marseille I, 2011. http://tel.archives-ouvertes.fr/tel-00684778.
Texto completoRichard, Mélissandre. "Application de la technique d'échange isotopique à l'étude de systèmes catalytiques innovants : activation et mobilité d'O2 sur YSZ au sein d’un double-lit et réactivité de l’azote dans les matériaux nitrures pour la catalyse hétérogène". Thesis, Poitiers, 2015. http://www.theses.fr/2015POIT2293/document.
Texto completoThis work concerns the study of new catalytic systems by isotopic exchange (IE) technique allowing to appreciate basic properties (molecules surface activation, mobility and reactivity of lattice atoms) to better understand catalytic mechanisms and to develop efficient catalysts. The identification of intermediate adsorbed species is possible by coupling mass spectrometry (gas-phase analysis) with the catalytic surface analysis by DRIFT spectroscopy.IE 16O/18O shows dispersal and synergetic effect of supported LaMnO3 perovskite (LM) on YSZ or TiO2 which explain catalytic performances of this perovskite structure for toluene oxidation via a suprafacial mechanism. IE C16O2/C18O2 activity demonstrates the remarkable lattice O atoms mobility of YSZ from 150 °C via adsorbed (hydrogeno)carbonates. To the contrary, in oxidation catalysis, under 800 °C, this mobility is very limited by O2 activation on YSZ surface. The solution proposed in this work is the previous generation of reactive oxygen species on a first catalytic bed of reducible material as LM. LM+YSZ dual-bed shows very efficient activity to reduce methane oxidation temperature at 425 °C via a Mars-van Krevelen (MvK) mechanism in which lattice O atoms of YSZ take part in the reaction by intermediate formate species.IE 14N/15N is thereafter used to analyse lattice N atoms reactivity of metal nitrides materials. In particular, Co3Mo3N and Ni2Mo3N show interesting properties depending on preparation or pre-treatement routes. This behaviour supposes that ammonia synthesis reaction could be procced via MvK type mechanism with the participation of lattice N of this nitrides
Petit, Jérôme. "CONTROLE DIMENSIONNEL SUB‐MICROMETRIQUE UTILISANT UN APPAREIL GONIOMETRIQUE BIDIMENSIONNEL RAPIDE Application à la microélectronique". Phd thesis, 2005. http://tel.archives-ouvertes.fr/tel-00635227.
Texto completoMangolini, Marc. "Apport de la fusion d'images satellitaires multicapteurs au niveau pixel en télédétection et photo-interprétation". Phd thesis, 1994. http://pastel.archives-ouvertes.fr/pastel-00957754.
Texto completo