Дисертації з теми "Détection du locuteur actif"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-24 дисертацій для дослідження на тему "Détection du locuteur actif".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Gacon, Pierre. "Analyse d'images et modèles de formes pour la détection et la reconnaissance : application aux visages en multimédia." Phd thesis, Grenoble INPG, 2006. http://tel.archives-ouvertes.fr/tel-00207391.
Повний текст джерелаDans ce travail, notre objectif est d'obtenir une détection robuste et efficace des contours des lèvres de façon à être capable de restaurer les mouvements de la parole aussi fidèlement que possible.
Nous apportons une attention particulière au contour intérieur de la bouche dans la segmentation est une tâche difficile à cause des variations non-linéaires de l'apparence.
Nous proposons une méthode basée sur un modèle statistique de la forme et de l'apparence échantillonnée faisant intervenir des descripteurs gaussiens locaux d'apparence.
Notre hypothèse est que la réponse de ces descripteurs locaux peut être prédite à partir de la forme par le biais d'un réseau de neurones non-linéaire.
Nous avons d'abord testé cette hypothèse dans un cas mono-locuteur et l'avons ensuite généralisé à un cas multi-locuteurs en tenant de la variabilité
inter-personne.
A cet effet, nous adaptons progressivement notre modèle au locuteur traité en déterminant son apparence caractéristique.
A partir de notre segmentation de la bouche, nous pouvons ensuite générer un clone de la bouche de la personne dont les mouvements seront aussi proches que possible de ceux de l'originale.
Finalement, nous avons évalué quantitativement puis qualitativement la pertinence de notre méthode en menant une expérience qui a quantifié l'apport effectif de compréhension de notre schéma d'analyse/synthèse dans le cas de numéros de téléphone en milieu bruité.
Pouthier, Baptiste. "Apprentissage profond et statistique sur données audiovisuelles dédié aux systèmes embarqués pour l'interface homme-machine." Electronic Thesis or Diss., Université Côte d'Azur, 2024. http://www.theses.fr/2024COAZ4019.
Повний текст джерелаIn the rapidly evolving landscape of human-machine interfaces, deep learning has been nothing short of revolutionary. It has ushered in a new era of audio-visual algorithms, which, in turn, have expanded the horizons of potential applications and strengthened the performance of traditional systems. However, these remarkable advancements come with a caveat - many of these algorithms are computationally demanding, rendering their integration onto embedded devices a formidable task. The primary focus of this thesis is to surmount this limitation through a comprehensive optimization effort, addressing the critical factors of latency and accuracy in audio-visual algorithms. Our approach entails a meticulous examination and enhancement of key components in the audio-visual human-machine interaction pipeline; we investigate and make contributions to fundamental aspects of audio-visual technology in Active Speaker Detection and Audio-visual Speech Recognition tasks. By tackling these critical building blocks, we aim to bridge the gap between the vast potential of audio-visual algorithms and their practical application in embedded systems. Our research introduces efficient models in Active Speaker Detection. On the one hand, our novel audio-visual fusion strategy yields significant improvements over other state-of-the-art systems, featuring a relatively simpler model. On the other hand, we explore neural architecture search, resulting in the development of a compact yet efficient architecture for the Active Speaker Detection problem. Furthermore, we present our work on audio-visual speech recognition, with a specific emphasis on keyword spotting. Our main contribution targets the visual aspect of speech recognition with a graph-based approach designed to streamline the visual processing pipeline, promising simpler audio-visual recognition systems
Kerkeni, Leila. "Analyse acoustique de la voix pour la détection des émotions du locuteur." Thesis, Le Mans, 2020. http://www.theses.fr/2020LEMA1003.
Повний текст джерелаThe aim of this thesis is to propose a speech emotion recognition (SER) system for application in classroom. This system has been built up using novel features based on the amplitude and frequency (AM-FM) modulation model of speech signal. This model is based on the joint use of empirical mode decomposition (EMD) and the Teager-Kaiser energy operator (TKEO). In this system, the discrete (or categorical) emotion theory was chosen to represent the six basic emotions (sadness, anger, joy, disgust, fear and surprise) and neutral emotion.Automatic recognition has been optimized by finding the best combination of features, selecting the most relevant ones and comparing different classification approaches. Two reference speech emotional databases, in German and Spanish, were used to train and evaluate this system. A new database in French, more appropriate for the educational context was built, tested andvalidated
François, Dominique. "Détection et identification des occlusives et fricatives au sein du système indépendant du locuteur APHODEX." Nancy 1, 1995. http://www.theses.fr/1995NAN10044.
Повний текст джерелаRemy, Bertrand. "Étude et réalisation d'un imageur actif laser à compression d'impulsion." Paris 11, 1986. http://www.theses.fr/1986PA112346.
Повний текст джерелаThe thesis describes the breadboard of an imaging CO₂ laser radar based upon the use of a CW waveguide single mode laser, heterodyne detection and chirp pulse compression. A theoretical model of the system is given: overall performance, angular, range and Doppler precisions are estimated, as well as the influence of heterodyne mixing, speckle, atmospheric turbulence and optical aberrations. An evaluation of several scanning devices is also given, that compares their performance as far as Doppler precision is concerned. Parametric optimization of this imaging system has 1ed to a total field of view of 0,5° x 1° (corresponding to 64 x 128 points). The 3 images Range-Doppler Velocity and Reflectivity are generated at a repetition rate of about 1,5 Hz. Field tests and experimental results confirm the theoretical investigations and help compare the relative interests of these new images in a wide range of applications
Ait, Fares Wassima. "Détection et suivi d'objets par vision fondés sur segmentation par contour actif basé région." Phd thesis, Université Paul Sabatier - Toulouse III, 2013. http://tel.archives-ouvertes.fr/tel-00932263.
Повний текст джерелаAit, Fares Wassima. "Détection et suivi d'objets par vision fondés sur segmentation par contour actif base région." Toulouse 3, 2013. http://thesesups.ups-tlse.fr/2143/.
Повний текст джерелаObject segmentation and tracking is a challenging area of ongoing research in computer vision. One important application lies in robotics where the ability to accurately segment an object of interest from its background is crucial and particularly on images acquired onboard during robot motion. Object segmentation technique consists in separating the object region from the image background according to a pre-defined criterion. Object tracking is a process of determining the positions of moving objects in image sequences. Several techniques can be applied to ensure these operations. In this thesis, we are interested to segment and track objects in video sequences using active contour method due to its robustness and efficiency to segment and track non-rigid objects. Active contour method consists in making a curve converge from an initial position around the object to be detected towards this object boundary according to a pre-defined criterion. We employ criteria which depend on the image regions what may impose certain constraints on the characteristics of these regions as a homogeneity assumption. This assumption may not always be verified due to the heterogeneity often present in images. In order to cope with the heterogeneity that may appear either in the object of interest or in the image background in noisy images using an inadequate active contour initialization, we propose a technique that combines local and global statistics in order to compute the segmentation criterion. By using a radius with a fixed size, a half-disk is superposed on each point of the active contour to define the local extraction regions. However, when the heterogeneity appears on both the object of interest and the image background, we develop a new technique based on a flexible radius that defines two half-disks with two different radius values to extract the local information. The choice of the value of these two radii is determined by taking into consideration the object size as well as the distance separating the object of interest from its neighbors. Finally, to track a mobile object within a video sequence using the active contour method, we develop a hybrid object tracking approach based on region characteristics and on motion vector of interest points extracted on the object region. Using our approach, the initial active contour for each image will be adequately adjusted in a way that it will be as close as possible to the actual boundary of the object of interest so that the evolution of active contour based on characteristics of the region will not be trapped by false contours. Simulation results on synthetic and real images validate the effectiveness of the proposed approaches
Cámara, Chávez Guillermo. "Analyse du contenu vidéo par apprentissage actif." Cergy-Pontoise, 2007. http://www.theses.fr/2007CERG0380.
Повний текст джерелаThis thesis presents work towards a unified framework for semi-automated video indexing and interactive retrieval. To create an efficient index, a set of representative key frames are selected from the entire video content. We developed an automatic shot boundary detection algorithm to get rid of parameters and thresholds. We adopted a SVM classifier due to its ability to use very high dimensional feature spaces while at the same time keeping strong generalization guarantees from few training examples. We deeply evaluated the combination of features and kernels and present interesting results obtained, for shot extraction TRECVID 2006 Task. We then propose an interactive video retrieval system: RETINVID, to significantly reduce the number of key frames annotated by the user. The key frames are selected based on their ability to increase the knowledge of the data. We perform an experiment against the 2005 TRECVID benchmark for high-level task
García-Gutiérrez, Luis Antonio. "Développement d'un contrôle actif tolérant aux défaillances appliqué aux systèmes PV." Thesis, Toulouse 3, 2019. http://www.theses.fr/2019TOU30071.
Повний текст джерелаThis work contributes by developing an active fault tolerant control (AFTC) for Photovoltaic (PV) systems. The fault detection and diagnosis (FDD) methodology is based on the analysis of a model that compares real-time measurement. We use a high granularity PV array model in the FDD tool to allow faults to be detected in complex conditions. Firstly, the research focuses on fault detection in complex shadow conditions. A real-time approach is presented to emulate the electrical characteristics of PV modules under complex shadow conditions. Using a precise emulators approach is a real challenge to study the high non-linearity and the complexity of PV systems in partial shading. The real-time emulation was validated with simple experimental results under failure conditions to design specific fault-detection algorithms in a first sample. The second part of the research addresses the FDD method for DC/DC and DC/AC power converters that are connected to the grid. Primary results allowed us to validate the system's recovery for normal operating points after a fault with this complete AFTC approach. Emulations based on the simulation of distributed power converters, fault detection methodologies based on a model, and a hybrid diagnostician were then presented
Motchon, Koffi Mawussé Djidula. "Caractérisation de la discernabilité des systèmes dynamiques linéaires et non-linéaires affines en la commande." Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10027/document.
Повний текст джерелаThe distinguishability of the input-output behavior of two dynamical systems plays a crucialrole in many applications such as control and observation of hybrid dynamical systems. Thisthesis aims to characterize this property of distinguishability. For linear systems and nonlinearcontrol-affine systems, we establish: conditions for strict distinguishability that ensure thedistinguishability of the systems for every control input jointly applied to them; conditions forcontrolled-distinguishability that guarantee the existence of a control input which makes distinguishable the outputs of the systems; conditions for residual-distinguishability that characterize the distinguishability of the modes through parity-space residuals. Moreover, in the linear case, a metric is specified in order to quantify for a given control input, the distinguishability degreeof the systems and the robustness of the property of distinguishability
Lelong, Amelie. "Convergence phonétique en interaction Phonetic convergence in interaction." Phd thesis, Université de Grenoble, 2012. http://tel.archives-ouvertes.fr/tel-00822871.
Повний текст джерелаBouguelia, Mohamed-Rafik. "Classification et apprentissage actif à partir d'un flux de données évolutif en présence d'étiquetage incertain." Thesis, Université de Lorraine, 2015. http://www.theses.fr/2015LORR0034/document.
Повний текст джерелаThis thesis focuses on machine learning for data classification. To reduce the labelling cost, active learning allows to query the class label of only some important instances from a human labeller.We propose a new uncertainty measure that characterizes the importance of data and improves the performance of active learning compared to the existing uncertainty measures. This measure determines the smallest instance weight to associate with new data, so that the classifier changes its prediction concerning this data. We then consider a setting where the data arrives continuously from an infinite length stream. We propose an adaptive uncertainty threshold that is suitable for active learning in the streaming setting and achieves a compromise between the number of classification errors and the number of required labels. The existing stream-based active learning methods are initialized with some labelled instances that cover all possible classes. However, in many applications, the evolving nature of the stream implies that new classes can appear at any time. We propose an effective method of active detection of novel classes in a multi-class data stream. This method incrementally maintains a feature space area which is covered by the known classes, and detects those instances that are self-similar and external to that area as novel classes. Finally, it is often difficult to get a completely reliable labelling because the human labeller is subject to labelling errors that reduce the performance of the learned classifier. This problem was solved by introducing a measure that reflects the degree of disagreement between the manually given class and the predicted class, and a new informativeness measure that expresses the necessity for a mislabelled instance to be re-labeled by an alternative labeller
Lelong, Amélie. "Convergence phonétique en interaction Phonetic convergence in interaction." Thesis, Grenoble, 2012. http://www.theses.fr/2012GRENT079/document.
Повний текст джерелаThe work presented in this manuscript is based on the study of a phenomenon called phonetic convergence which postulates that two people in interaction will tend to adapt how they talk to their partner in a communicative purpose. We have developed a paradigm called “Verbal Dominoes“ to collect a large corpus to characterize this phenomenon, the ultimate goal being to fill a conversational agent of this adaptability in order to improve the quality of human-machine interactions.We have done several studies to investigate the phenomenon between pairs of unknown people, good friends, and between people coming from the same family. We expect that the amplitude of convergence is proportional to the social distance between the two speakers. We found this result. Then, we have studied the knowledge of the linguistic target impact on adaptation. To characterize the phonetic convergence, we have developed two methods: the first one is based on a linear discriminant analysis between the MFCC coefficients of each speaker and the second one used speech recognition techniques. The last method will allow us to study the phenomenon in less controlled conditions.Finally, we characterized the phonetic convergence with a subjective measurement using a new perceptual test called speaker switching. The test was performed using signals coming from real interactions but also with synthetic data obtained with the harmonic plus
Bouguelia, Mohamed-Rafik. "Classification et apprentissage actif à partir d'un flux de données évolutif en présence d'étiquetage incertain." Electronic Thesis or Diss., Université de Lorraine, 2015. http://www.theses.fr/2015LORR0034.
Повний текст джерелаThis thesis focuses on machine learning for data classification. To reduce the labelling cost, active learning allows to query the class label of only some important instances from a human labeller.We propose a new uncertainty measure that characterizes the importance of data and improves the performance of active learning compared to the existing uncertainty measures. This measure determines the smallest instance weight to associate with new data, so that the classifier changes its prediction concerning this data. We then consider a setting where the data arrives continuously from an infinite length stream. We propose an adaptive uncertainty threshold that is suitable for active learning in the streaming setting and achieves a compromise between the number of classification errors and the number of required labels. The existing stream-based active learning methods are initialized with some labelled instances that cover all possible classes. However, in many applications, the evolving nature of the stream implies that new classes can appear at any time. We propose an effective method of active detection of novel classes in a multi-class data stream. This method incrementally maintains a feature space area which is covered by the known classes, and detects those instances that are self-similar and external to that area as novel classes. Finally, it is often difficult to get a completely reliable labelling because the human labeller is subject to labelling errors that reduce the performance of the learned classifier. This problem was solved by introducing a measure that reflects the degree of disagreement between the manually given class and the predicted class, and a new informativeness measure that expresses the necessity for a mislabelled instance to be re-labeled by an alternative labeller
Diou, Odile. "Synthèse de nanocapsules polymères pour la détection de tumeurs solides par échographie et IRM du Fluor : vers un outil théranostique." Phd thesis, Université Paris Sud - Paris XI, 2012. http://tel.archives-ouvertes.fr/tel-00907145.
Повний текст джерелаBendris, Meriem. "Indexation audio-visuelle des personnes dans un contexte de télévision." Phd thesis, Paris, Télécom ParisTech, 2011. https://pastel.hal.science/pastel-00661662.
Повний текст джерелаWith increasing internet use, the amount of multimedia content multiplies, making it necessary to develop technologies in order to enable users to browse through the multimedia data. One key element for browsing is the presence of people. However, structuring TV-Content in terms of people is a hard problem due to many difficulties in audio and visual modalities as well as in their association (short speaker turns, variations in facial expressions and pose, no synchronization between sequences of a person's appearance and sequences of his/her speech). The goal underlying this dissertation is to structure TV-Content by person in order to allow users to navigate through sequences in which a particular individual appears. To this end, most methods propose indexing people separately by the audio and visual information and then associating the results of each in order to obtain a talking-face index. Unfortunately, this type of approach combines clustering errors present in each modality. Our work seeks to capitalise on interactions between the audio and visual modalities rather than treating them separately. We propose a mutual correction scheme for audio and visual clustering errors. First, the clustering errors are detected using indicators that suspect a talking-face presence (Step 1). Then, the incorrect label is corrected according to an automatic modification scheme (Step 2). In more detail, first we proposed a Baseline system of talking faces indexing in which audio and visual indexes of people are generated independently by speaker and clothes clustering. Then, we proposed a fusion method based on maximizing global coverage of detected clusters. Results on a TV-show database show a high precision (90%), but with a significant missed-detection rate (only 55% of talking faces sequences are detected). In order to automatically detect a talking face presence (in the step 1), we exploited the fact that the lip-activity is strongly related to speech activity. We developed a new method for lip-activity detection in TV-Context based on the disorder of the directions of pixels. An evaluation is performed on manually annotated TV-Shows and significant improvement is observed compared to the state-of-the-art in TV-Contexts. Next, the modification method is based on the paradigm that one modality (either audio or visual) is more reliable than the other. We proposed two modification schemes: one based on systematic correction of the supposedly less reliable modality a priori while the second proposes comparing unsupervised audio-visual model scores to determine which modality failed. The unsupervised models are trained from the homogeneous sets of talking faces obtained automatically by the Baseline system. Experiments conducted on a TV-show database show that the proposed correction schemes yield significant improvement in performance, mainly due to an important reduction of missed talking-faces. We have investigated also on late fusion techniques for identity verification in biometric systems. We have proposed a fusion method based on the signal quality in each modality
Peng, Kun. "Compression et vérification de visages pour les cartes d'identité à puce." Caen, 2009. http://www.theses.fr/2009CAEN2060.
Повний текст джерелаSmart identification cards will undoubtedly replace the current paper-based ones and become the next generation of identification cards. This Ph. D. Thesis presents and proposes a smart identification card system using facial biometric information for identity authentification on following a strategy containing three security levels: manual verification offline, manual verification online and automatic verification. The principal task of manual face verification online is to find out a compact and secure way to stock a photograph of the identification card holder into the chip on the smart card. We bring out an idea based on decomposing the face image into two parts which are stocked into the card and the database of system respectively. An algorithm based on Singular Value Decomposition of 2D matrix is proposed to realize this idea. The work on automatic face verification faces mainly to establish a reliable face verification algorithm adapting to the context. A novel face verification scheme based on class-specific face models is proposed in this thesis. And the technique Active Appearance Model (AAM) is applied for face modeling and representation. To improve the performance of the system, we developed an algorithm of eye position detection, which can offer a better initialization for AAM matching, and a technique of virtual faces generation, which can enrich the variety of class-specific AAM models. In the sight of the performance evaluation of the proposed method, the database XM2VTS and the accompanied protocol, Lausanne Protocol, are selected to make the experiments
Bendris, Meriem. "Indexation audio-visuelle des personnes dans un contexte de télévision." Phd thesis, Télécom ParisTech, 2011. http://pastel.archives-ouvertes.fr/pastel-00661662.
Повний текст джерелаDaloy, Patrick. "RU 47213, cholinomimétique : étude bibliographique du parachlorophénol libéré au cours de la biotransformation du RU 47213 en son métabolite actif, le RU 35963. Validation d'une méthode de dosage du RU 35963 par HPLC et détection UV." Paris 5, 1995. http://www.theses.fr/1995PA05P242.
Повний текст джерелаBeaugnon, Anaël. "Expert-in-the-loop supervised learning for computer security detection systems." Thesis, Paris Sciences et Lettres (ComUE), 2018. http://www.theses.fr/2018PSLEE049/document.
Повний текст джерелаThe overall objective of this thesis is to foster the deployment of supervised learning in detection systems to strengthen detection. To that end, we consider the whole machine learning pipeline (data annotation, feature extraction, training, and evaluation) with security experts as its core since it is crucial to pursue real-world impact. First, we provide methodological guidance to help security experts build supervised detection models that suit their operational constraints. Moreover, we design and implement DIADEM, an interactive visualization tool that helps security experts apply the methodology set out. DIADEM deals with the machine learning machinery to let security experts focus mainly on detection. Besides, we propose a solution to effectively reduce the labeling cost in computer security annotation projects. We design and implement an end-to-end active learning system, ILAB, tailored to security experts needs. Our user experiments on a real-world annotation project demonstrate that they can annotate a dataset with a low workload thanks to ILAB. Finally, we consider automatic feature generation as a means to ease, and thus foster, the use of machine learning in detection systems. We define the constraints that such methods should meet to be effective in building detection models. We compare three state-of-the-art methods based on these criteria, and we point out some avenues of research to better tailor automatic feature generation to computer security experts needs
Guigou, Fabio. "The artificial immune ecosystem : a scalable immune-inspired active classifier, an application to streaming time series analysis for network monitoring." Thesis, Strasbourg, 2019. http://www.theses.fr/2019STRAD007/document.
Повний текст джерелаSince the early 1990s, immune-inspired algorithms have tried to adapt the properties of the biological immune system to various computer science problems, not only in computer security but also in optimization and classification. This work explores a different direction for artificial immune systems, focussing on the interaction between subsystems rather than the biological processes involved in each one. These patterns of interaction in turn create the properties expected from immune systems, namely their ability to detect anomalies, memorize their signature to react quickly upon secondary exposure, and remain tolerant to symbiotic foreign organisms such as the intestinal fauna. We refer to a set of interacting systems as an ecosystem, thus this new approach has called the Artificial Immune Ecosystem. We demonstrate this model in the context of a real-world problem where scalability and performance are essential: network monitoring. This entails time series analysis in real time with an expert in the loop, i.e. active learning instead of supervised learning
Bayoudh, Mehdi. "Active Diagnosis of Hybrid Systems Guided by Diagnosability Properties - Application to Autonomous Satellites." Thesis, Toulouse, INPT, 2009. http://www.theses.fr/2009INPT069H.
Повний текст джерелаMotivated by the requirements of the space domain in terms of on-board diagnosis and autonomy, this thesis addresses the problems of diagnosis, diagnosability and active diagnosis of hybrid systems. Supported by a hybrid modeling framework, a passive approach for model-based diagnosis mixing discrete-event and continuous techniques is proposed. The same hybrid model is used to define the diagnosability property for hybrid systems and diagnosability criteria are derived. When the diagnosis provided by the passive diagnosis approach is ambiguous, active diagnosis is needed. This work provides a method for performing such active diagnosis. Starting with an ambiguous belief state, the method calls for diagnosability analysis results to determine a new system configuration in which fault candidates can be discriminated. Based on a new finite state machine called the diagnoser, the active diagnosis is formulated as a conditional planning problem and an AND-OR graph exploration algorithm is proposed to determine active diagnosis plans. Finally, the diagnosis approach is tested on the Attitude Control System (ACS) of a satellite simulator provided by Thales Alenia Space. The diagnosis module is successfully tested on several fault scenarios and the obtained results are reported
Xu, Yanli. "Une mesure de non-stationnarité générale : Application en traitement d'images et du signaux biomédicaux." Thesis, Lyon, INSA, 2013. http://www.theses.fr/2013ISAL0090/document.
Повний текст джерелаThe intensity variation is often used in signal or image processing algorithms after being quantified by a measurement method. The method for measuring and quantifying the intensity variation is called a « change measure », which is commonly used in methods for signal change detection, image edge detection, edge-based segmentation models, feature-preserving smoothing, etc. In these methods, the « change measure » plays such an important role that their performances are greatly affected by the result of the measurement of changes. The existing « change measures » may provide inaccurate information on changes, while processing biomedical images or signals, due to the high noise level or the strong randomness of the signals. This leads to various undesirable phenomena in the results of such methods. On the other hand, new medical imaging techniques bring out new data types and require new change measures. How to robustly measure changes in theos tensor-valued data becomes a new problem in image and signal processing. In this context, a « change measure », called the Non-Stationarity Measure (NSM), is improved and extended to become a general and robust « change measure » able to quantify changes existing in multidimensional data of different types, regarding different statistical parameters. A NSM-based change detection method and a NSM-based edge detection method are proposed and respectively applied to detect changes in ECG and EEG signals, and to detect edges in the cardiac diffusion weighted (DW) images. Experimental results show that the NSM-based detection methods can provide more accurate positions of change points and edges and can effectively reduce false detections. A NSM-based geometric active contour (NSM-GAC) model is proposed and applied to segment the ultrasound images of the carotid. Experimental results show that the NSM-GAC model provides better segmentation results with less iterations that comparative methods and can reduce false contours and leakages. Last and more important, a new feature-preserving smoothing approach called « Nonstationarity adaptive filtering (NAF) » is proposed and applied to enhance human cardiac DW images. Experimental results show that the proposed method achieves a better compromise between the smoothness of the homogeneous regions and the preservation of desirable features such as boundaries, thus leading to homogeneously consistent tensor fields and consequently a more reconstruction of the coherent fibers
Dehaes, Mathieu. "Représentations analytiques des objets géométriques et contours actifs en imagerie." Thèse, 2004. http://hdl.handle.net/1866/14589.
Повний текст джерела