Gotowa bibliografia na temat „Mélange de gaussiennes photométriques”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Spis treści
Zobacz listy aktualnych artykułów, książek, rozpraw, streszczeń i innych źródeł naukowych na temat „Mélange de gaussiennes photométriques”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Rozprawy doktorskie na temat "Mélange de gaussiennes photométriques"
Guerbas, Seif Eddine. "Modélisation adaptée des images omnidirectionnelles pour agrandir le domaine de convergence de l'asservissement visuel virtuel direct". Electronic Thesis or Diss., Amiens, 2022. http://www.theses.fr/2022AMIE0026.
Pełny tekst źródłaOmnidirectional vision captures a scene in real-time in all directions with a wider field of view than a conventional camera. Within the environment, linking the visual features contained in the camera images to its movements is a central issue for visual servoing. Direct approaches, however, are characterized by a limited range of convergence. The main objective of this dissertation is to significantly extend the area of convergence in the context of virtual visual servoing by representing the omnidirectional image by a Photometric Gaussian Mixtures (PGM). This approach is further extended in the second step to the registration and direct tracking based on 3D models in omnidirectional images. This proposed methodology allows for studying the localization of a mobile robot equipped with a panoramic camera in a 3D urban model. The results show a significant enlargement of the convergence domain for high robustness to large interframe movements, as evidenced by experiments in virtual environments and with real images captured with a mobile robot and a vehicle
Habibi, Zaynab. "Vers l'assistance à l'exploration pertinente et réaliste d'environnements 3D très denses". Thesis, Amiens, 2015. http://www.theses.fr/2015AMIE0028/document.
Pełny tekst źródłaIn this thesis, we address the issue of navigation in virtual 3D environment. In particular, environments made of hundreds of millions of points, which are difficult to bring under control by a novice. The complexity and the wealth of details of the 3D point cloud of the cathedral of Amiens can result in a disorientation and in an irrelevant visualization with existing tools (interfaces). The contributions of the thesis deal with automatic or assisted camera control exploiting 2D visual information from the image and other 3D information from the environment. To ensure the visual relevance, we propose two methods to pilot the camera, one based on the photometric entropy and the second representing the major contribution of this thesis, defines and exploits the saliency-based Gaussian mixture. The visual servoing formalism is used to link the image modelling to the camera degrees of freedom. The obstacle avoidance, the fluidity of motion and appropriate camera orientation are considered as additional constraints taken into account in two navigation modes: the local framing and the global exploration. The goal of visual framing is to move the camera by maximizing the saliency-based Gaussian mixture feature, in order to reach a relevant viewpoint to visualize an object. We test this approach in synthetic model, 3D points cloud model and in a real environment with a robot. Regarding exploration, we present first an automatic camera control exploiting the photometric entropy and some constraints to ensure realistic motion. The problem is solved using an hybrid and hierarchical optimization algorithm. Then, we present a navigation aid system helping the user to explore a part or the whole 3D environment. The system is built using the redundancy formalism taking into account several constraints. These approaches were tested on simple and complex dense 3D points cloud
Darwich, Ali. "Approche pixel de la soustraction d'arrière-plan en vidéo, basée sur un mélange de gaussiennes imprécises". Thesis, Littoral, 2018. http://www.theses.fr/2018DUNK0479/document.
Pełny tekst źródłaMoving objects detection is a very important step for many applications such as human behavior analysis surveillance, model-based action recognition, road traffic monitoring, etc. Background subtraction is a popular approach, but difficult given that it must overcome many obstacles, such as dynamic background changes, brightness variations, occlusions, and so on. In the presented works, we focused on this problem of objects/background segmentation, using a type-2 fuzzy modeling to manage the inaccuracy of the model and the data. The proposed method models the state of each pixel using an imprecise and scalable Gaussian mixture model, which is exploited by several fuzzy classifiers to ultimately estimate the pixel class at each image. More precisely, this decision takes into account the history of its evolution, but also its spatial neighborhood and its possible displacements in the preceding images. Then we compared the proposed method with other close methods, including methods based on a gaussian mixture model, fuzzy based methods, or ACP type methods. This comparison allowed us to assess its good performances, and to propose some perspectives to this work
Genin, Laure. "Détection d'objets de petite taille sur des séquences aériennes ou satellitaires". Paris 13, 2013. http://scbd-sto.univ-paris13.fr/secure/edgalilee_th_2013_genin.pdf.
Pełny tekst źródłaThe objective of this thesis is to improve the detection of point objects in optical imaging. They focus on the challenging detection of low velocity point objects on inhomogeneous background including areas of strong gradients of gray levels. In this context, we propose single-frame detection methods trying to take advantage at best of the spatial background correlation. Spatio-temporal extensions of the proposed methods are studied in a second time. Based on a formalism of the generalized likelihood ratio test (GLRT), the problem of detection boils down to a two-step process which consists in separating the first and second order estimation of the local background (i. E. Mean and covariance). To improve the performances of the detection methods by first order background modelling, we adapt patch-based denoising method to detection. Despite the improvement of detection performance brought by these patch-based methods, it appears that textures associated with background structures are still visible after the background suppression step. We seek to improve the detection performance by second order modeling. We are interested in matched filter adapted by area based on a Gaussian mixture model. A detailed performance analysis of the developed filters is made from real cloudy background on which point targets are embedded
Zaïdi, Abdelhamid. "Séparation aveugle d'un mélange instantané de sources autorégressives gaussiennes par la méthode du maximum de vraissemblance exact". Université Joseph Fourier (Grenoble), 2000. http://www.theses.fr/2000GRE10233.
Pełny tekst źródłaJarraya, Siala Aida. "Nouvelles paramétrisations de réseaux bayésiens et leur estimation implicite : famille exponentielle naturelle et mélange infini de Gaussiennes". Phd thesis, Nantes, 2013. https://archive.bu.univ-nantes.fr/pollux/show/show?id=aef89743-c009-457d-8c27-a888655a4e58.
Pełny tekst źródłaLearning a Bayesian network consists in estimating the graph (structure) and the parameters of conditional probability distributions associated with this graph. Bayesian networks learning algorithms rely on classical Bayesian estimation approach whose a priori parameters are often determined by an expert or defined uniformly The core of this work concerns the application of several advances in the field of statistics as implicit estimation, Natural exponential families or infinite mixtures of Gaussian in order to (1) provide new parametric forms for Bayesian networks, (2) estimate the parameters of such models and (3) learn their structure
Jarraya, Siala Aida. "Nouvelles paramétrisations de réseaux Bayésiens et leur estimation implicite - Famille exponentielle naturelle et mélange infini de Gaussiennes". Phd thesis, Université de Nantes, 2013. http://tel.archives-ouvertes.fr/tel-00932447.
Pełny tekst źródłaDumitru, Corneliu Octavian. "Noise sources in robust uncompressed video watermarking". Phd thesis, Institut National des Télécommunications, 2010. http://tel.archives-ouvertes.fr/tel-00541755.
Pełny tekst źródłaDevinck, Vincent. "Systèmes dynamiques linéaires : vitesse de mélange et spectre ponctuel unimodulaire". Thesis, Lille 1, 2012. http://www.theses.fr/2012LIL10006/document.
Pełny tekst źródłaIn this thesis, we study into two different parts the eigenvectors associated to unimodular eigenvalues of an operator on a separable Banach space. The first part of the thesis follows a work of F. Bayart and S. Grivaux where they give condition on the eigenvectors associated to unimodular eigenvalues of an operator on a complex separable Hilbert space to admit a Gaussian measure for which the operator defines a strongly mixing transformation. With this condition on the eigenvectors, we investigate the subject of speed of mixing of the strongly mixing operator. We prove that there is no way to obtain a uniform speed of mixing in general. Then we prove that if the eigenvectors associated to unimodular eigenvalues of the operator are parametrized by a countable family of regular eigenvector fields then we have a speed of mixing by considering regular classes of functions. In the second part of the thesis, we study the unimodular point spectrum of an operator on a separable Banach space. By using the results on Jamison sequences, we give a characterization of Jamison sequences for strongly continuous semigroups. We are also concerned in the problem of construction of Banach space and operator on this space when the sequences are not Jamison sequences. Then we generalize the notion of Jamison sequence by studying the unimodular point spectrum of a group representation which is bounded with respect to some sequence of this group. In particular, we characterize Jamison sequences of a finitely generated abelian group
Fernandes, maligo Artur otavio. "Unsupervised Gaussian mixture models for the classification of outdoor environments using 3D terrestrial lidar data". Thesis, Toulouse, INSA, 2016. http://www.theses.fr/2016ISAT0053/document.
Pełny tekst źródłaThe processing of 3D lidar point clouds enable terrestrial autonomous mobile robots to build semantic models of the outdoor environments in which they operate. Such models are interesting because they encode qualitative information, and thus provide to a robot the ability to reason at a higher level of abstraction. At the core of a semantic modelling system, lies the capacity to classify the sensor observations. We propose a two-layer classi- fication model which strongly relies on unsupervised learning. The first, intermediary layer consists of a Gaussian mixture model. This model is determined in a training step in an unsupervised manner, and defines a set of intermediary classes which is a fine-partitioned representation of the environment. The second, final layer consists of a grouping of the intermediary classes into final classes that are interpretable in a considered target task. This grouping is determined by an expert during the training step, in a process which is supervised, yet guided by the intermediary classes. The evaluation is done for two datasets acquired with different lidars and possessing different characteristics. It is done quantitatively using one of the datasets, and qualitatively using another. The system is designed following the standard learning procedure, based on a training, a validation and a test steps. The operation follows a standard classification pipeline. The system is simple, with no requirement of pre-processing or post-processing stages