Publications de type 'techreport'

Résultat de la recherche dans la liste des publications :

• 90 Rapports de recherche et Rapports techniques

nouvelle recherche | bibtex

navigation : 1 - 2 - 3- ... - 8 - 9 affichage : 10 par page | 20 par page | 50 par page

90 Rapports de recherche et Rapports techniques

11 - Modeling the statistics of high resolution SAR images. V. Krylov et G. Moser et S.B. Serpico et J. Zerubia. Research Report 6722, INRIA, novembre 2008. Mots-clés : Synthetic Aperture Radar (SAR) image, Probability density function, parametric estimation, finite mixture models, EM Stochastique (SEM). Copyright : INRIA/ARIANA, 2008 @TECHREPORT{krylovDSEM08, author = {Krylov, V. and Moser, G. and Serpico, S.B. and Zerubia, J.}, title = {Modeling the statistics of high resolution SAR images}, year = {2008}, month = {novembre}, institution = {INRIA}, type = {Research Report}, number = {6722}, url = {http://hal.archives-ouvertes.fr/inria-00342681/en/}, pdf = {http://hal.archives-ouvertes.fr/docs/00/35/76/27/PDF/RR-6722.pdf}, keyword = {Synthetic Aperture Radar (SAR) image, Probability density function, parametric estimation, finite mixture models, EM Stochastique (SEM)} } Abstract : In the context of remotely sensed data analysis, a crucial problem is represented by the need to develop accurate models for the statistics of pixel intensities. In this work, we develop a parametric finite mixture model for modelling the statistics of intensities in high resolution Synthetic Aperture Radar (SAR) images. Along with the models we design an efficient parameter estimation scheme by integrating the Stochastic Expectation Maximization scheme and the Method of log-cumulants with an automatic technique to select, for each mixture component, an optimal parametric model taken from a predefined dictionary of parametric probability density functions (pdf). In particular, the proposed dictionary consists of eight most efficient state-of-the-art SAR-specific pdfs: Nakagami, log-normal, generalized Gaussian Rayleigh, Heavy-tailed Rayleigh, Weibull, K-root, Fisher and generalized Gamma. The experiment results with a set of several real SAR (COSMO-SkyMed) images demonstrate the high accuracy of the designed algorithm, both from the viewpoint of a visual comparison of the histograms, and from the viewpoint of quantitive measures such as correlation coefficient (always above 99,5%) . We stress, in particular, that the method proves to be effective on all the considered images, remaining accurate for multimodal and highly heterogeneous images.

12 - Parametric blind deconvolution for confocal laser scanning microscopy-proof of concept. P. Pankajakshan et L. Blanc-Féraud et B. Zhang et Z. Kam et J.C. Olivo-Marin et J. Zerubia. Rapport de Recherche 6493, INRIA, avril 2008. Mots-clés : Confocal Laser Scanning Microscopy, Bayesian restoration, Blind Deconvolution, point spread function, Richardson-Lucy algorithm, Variation totale. Copyright : ARIANA/INRIA @TECHREPORT{ppankajakshan08b, author = {Pankajakshan, P. and Blanc-Féraud, L. and Zhang, B. and Kam, Z. and Olivo-Marin, J.C. and Zerubia, J.}, title = {Parametric blind deconvolution for confocal laser scanning microscopy-proof of concept}, year = {2008}, month = {avril}, institution = {INRIA}, type = {Research Report}, number = {6493}, url = {https://hal.inria.fr/inria-00269265}, pdf = {http://hal.inria.fr/docs/00/27/02/92/PDF/report.pdf}, keyword = {Confocal Laser Scanning Microscopy, Bayesian restoration, Blind Deconvolution, point spread function, Richardson-Lucy algorithm, Variation totale} } Résumé : Nous proposons une méthode de restauration itérative d’images de fluorescence CLSM et d’estimation paramétrique de la fonction de flou (PSF) du système d’acquisition. Le CLSM est un microscope qui balaye un échantillon en 3D et utilise une sténopée pour rejeter la lumière en dehors du point de focalisation. Néanmoins, la qualité des images souffre de deux limitations physiques. La première est due à la diffraction due au système optique et la seconde est due à la quantité réduite de lumière détectée par le tube photo-multiplicateur (PMT). Ces limitations induisent respectivement un flou et du bruit de comptage de photons. Les images peuvent alors bénéficier d’un post-traitement de restauration fondé sur la déconvolution. Le problème à traiter est l’estimation simultanée de la distribution 3D de l’échantillon des sources fluorescentes et de la PSF du microscope (i.e. de déconvolution aveugle). En utilisant un modèle de processus physique d’acquisition d’images microscopiques (CLSM), on réduit le nombre de paramètres libres décrivant la PSF et on introduit des contraintes. On introduit aussi des connaissances a priori sur l’échantillon ce qui permet de stabiliser le processus d’estimation et de favoriser la convergence. Des expériences sur des données synthétiques montrent que la PSF peut être estimée avec précision. Des expériences sur des données réelles montrent de bons resultats de déconvolution en comparaison avec le modèle théorique de la PSF du microscope. Abstract : We propose a method for the iterative restoration of fluorescence Confocal Laser Scanning Microscope (CLSM) images with parametric estimation of the acquisition system’s Point Spread Function (PSF). The CLSM is an optical fluorescence microscope that scans a specimen in 3D and uses a pinhole to reject most of the out-of-focus light. However, the quality of the image suffers from two primary physical limitations. The first is due to the diffraction-limited nature of the optical system and the second is due to the reduced amount of light detected by the photomultiplier tube (PMT). These limitations cause blur and photon counting noise respectively. The images can hence benefit from post-processing restoration methods based on deconvolution. An efficient method for parametric blind image deconvolution involves the simultaneous estimation of the specimen 3D distribution of fluorescent sources and the microscope PSF. By using a model for the microscope image acquisition physical process, we reduce the number of free parameters describing the PSF and introduce constraints. The parameters of the PSF may vary during the course of experimentation, and so they have to be estimated directly from the observation data. We also introduce a priori knowledge of the specimen that permits stabilization of the estimation process and favorizes the convergence. Experiments on simulated data show that the PSF could be estimatedwith a higher degree of accuracy and those done on real data show very good deconvolution results in comparison to the theoretical microscope PSF model.

13 - On the illumination invariance of the level lines under directed light. Application to change detection. P. Weiss et A. Fournier et L. Blanc-Féraud et G. Aubert. Rapport de Recherche 6612, INRIA, 2008. Mots-clés : Level Lines, illumination invariance, topographic map, Change detection, remote sensing, Zones urbaines. Copyright : @TECHREPORT{RR-6612, author = {Weiss, P. and Fournier, A. and Blanc-Féraud, L. and Aubert, G.}, title = {On the illumination invariance of the level lines under directed light. Application to change detection}, year = {2008}, institution = {INRIA}, type = {Research Report}, number = {6612}, url = {https://hal.archives-ouvertes.fr/inria-00310383}, pdf = {http://hal.inria.fr/docs/00/31/03/83/PDF/RR-6612.pdf}, keyword = {Level Lines, illumination invariance, topographic map, Change detection, remote sensing, Zones urbaines} } Abstract : We analyze the illumination invariance of the level lines of an image. We show that if the scene surface has Lambertian reflectance and the light is directed, then a necessary condition for the level lines to be illumination invariant is that the 3D scene be developable and that its albedo satisfies some geometrical constraints. We then show that the level lines are ``almost'' invariant for piecewise developable surfaces. Such surfaces fit most of the urban structures. In a second part, this allows us to devise a very fast algorithm that detects changes between pairs of remotely sensed images of urban areas, independently of the lighting conditions. We show the effectiveness of the algorithm both on synthetic OpenGL scenes and real Quickbird images. We compare the efficiency of the proposed algorithm with other classical approaches and show that it is superior both in practice and in theory.

14 - Reconstruction d'images satellitaires à partir d'un échantillonnage irrégulier. M. Carlavan et P. Weiss et L. Blanc-Féraud et J. Zerubia. Rapport de Recherche 6732, INRIA, 2008. Mots-clés : l1 norm, nesterov scheme, total variation minimization, wavelet. Copyright : @TECHREPORT{RR-6732, author = {Carlavan, M. and Weiss, P. and Blanc-Féraud, L. and Zerubia, J.}, title = {Reconstruction d'images satellitaires à partir d'un échantillonnage irrégulier}, year = {2008}, institution = {INRIA}, type = {Research Report}, number = {6732}, url = {http://hal.archives-ouvertes.fr/inria-00340975/fr/}, pdf = {http://hal.inria.fr/docs/00/34/09/75/PDF/RR-6732.pdf}, keyword = {l1 norm, nesterov scheme, total variation minimization, wavelet} }

15 - Support Vector Machines for burnt area discrimination. O. Zammit et X. Descombes et J. Zerubia. Rapport de Recherche 6343, INRIA, novembre 2007. Mots-clés : Feux de foret, Zones brûlées, Imagerie satellitaire, Support Vector Machines, Classification. @TECHREPORT{zammit_RR_07, author = {Zammit, O. and Descombes, X. and Zerubia, J.}, title = {Support Vector Machines for burnt area discrimination}, year = {2007}, month = {novembre}, institution = {INRIA}, type = {Research Report}, number = {6343}, url = {http://hal.inria.fr/inria-00185101/fr/}, pdf = {http://hal.inria.fr/inria-00185101/fr/}, keyword = {Feux de foret, Zones brûlées, Imagerie satellitaire, Support Vector Machines, Classification} } Résumé : Ce rapport aborde le problème de l'évaluation des dégâts après un feux de forêt. La détection est effectuée à partir d'une seule image satellite (SPOT 5) acquise après le feu. Afin de détecter les zones brûlées, nous utilisons une approche récente de classification nommée SVM (Séparateurs à Vaste Marge). Cette méthode est comparée aux algorithmes de classification plus conventionnels comme les K-moyennes ou les K-plus proches voisins, qui sont régulièrement utilisés en traitement d'image. Nous proposons également une méthode de classification non supervisée combinant les K-moyennes et les SVM. Les résultats fournis par les différentes techniques sont comparés à des vérités de terrain sur diverses zones brûlées. Abstract : This report addresses the problem of burnt area discrimination using remote sensing images. The detection is based on a single post-fire image acquired by SPOT 5 satellite. To delineate the burnt areas, we use a recent classification method called Support Vectors Machines (SVM). This approach is compared to more conventional classifiers such as K-means or K-nearest neighbours which are widely used in image processing. We also proposed a new automatic classification approach combining K-means and SVM. The results given by the different methods are finally compared to ground truths on various burnt areas

16 - Détection de flamants roses par processus ponctuels marqués pour l'estimation de la taille des populations. S. Descamps et X. Descombes et A. Béchet et J. Zerubia. Research Report 6328, INRIA, octobre 2007. Mots-clés : Extraction d'objets, modélisation stochastique , Processus ponctuels marques, dynamique de naissance/mort, environnement, flamants roses. @TECHREPORT{Descamps-Descombes, author = {Descamps, S. and Descombes, X. and Béchet, A. and Zerubia, J.}, title = {Détection de flamants roses par processus ponctuels marqués pour l'estimation de la taille des populations}, year = {2007}, month = {octobre}, institution = {INRIA}, type = {Research Report}, number = {6328}, url = {http://hal.inria.fr/inria-00180811}, pdf = {http://hal.inria.fr/docs/00/18/08/93/PDF/RR-Desc-Desc-Bech-Zeru.pdf}, keyword = {Extraction d'objets, modélisation stochastique , Processus ponctuels marques, dynamique de naissance/mort, environnement, flamants roses} }

17 - An adaptive simulated annealing cooling schedule for object detection in images. M. Ortner et X. Descombes et J. Zerubia. Rapport de Recherche 6336, INRIA, octobre 2007. Mots-clés : Traitement d'image, Shape extraction, Spatial point process, Recuit Simule, Adaptive cooling schedule. @TECHREPORT{Ortner-Descombes, author = {Ortner, M. and Descombes, X. and Zerubia, J.}, title = {An adaptive simulated annealing cooling schedule for object detection in images}, year = {2007}, month = {octobre}, institution = {INRIA}, type = {Research Report}, number = {6336}, url = {https://hal.inria.fr/inria-00181764}, pdf = {https://hal.inria.fr/inria-00181764}, keyword = {Traitement d'image, Shape extraction, Spatial point process, Recuit Simule, Adaptive cooling schedule} }

18 - Efficient schemes for total variation minimization under constraints in image processing. P. Weiss et L. Blanc-Féraud et G. Aubert. Rapport de Recherche 6260, INRIA, juillet 2007. Mots-clés : l1 norm, total variation minimization, duality lp norms, gradient and subgradient descent, nesterov scheme, texture + geometry decomposition. @TECHREPORT{RR-6260, author = {Weiss, P. and Blanc-Féraud, L. and Aubert, G.}, title = {Efficient schemes for total variation minimization under constraints in image processing}, year = {2007}, month = {juillet}, institution = {INRIA}, type = {Research Report}, number = {6260}, url = {http://hal.inria.fr/inria-00166096/fr/}, pdf = {http://hal.inria.fr/docs/00/26/16/35/PDF/RR-6260.pdf}, ps = {http://hal.inria.fr/docs/00/26/16/35/PS/RR-6260.ps}, keyword = {l1 norm, total variation minimization, duality lp norms, gradient and subgradient descent, nesterov scheme, texture + geometry decomposition} } Résumé : Ce papier présente de nouveaux algorithmes pour minimiser la variation totale, et plus généralement des normes l^1, sous des contraintes convexes. Ces algorithmes proviennent d'une avancée récente en optimisation convexe proposée par Yurii Nesterov. Suivant la régularité de l'attache aux données, nous résolvons soit un problème primal, soit un problème dual. Premièrement, nous montrons que les schémas standard de premier ordre permettent d'obtenir des solutions de précision epsilon en O(frac1epsilon^2) itérations au pire des cas. Pour une contrainte convexe quelconque, nous proposons un schéma qui permet d'obtenir une solution de précision epsilon en O(frac1epsilon) itérations. Pour une contrainte fortement convexe, nous résolvons un problème dual avec un schéma qui demande O(frac1sqrtepsilon) itérations pour obtenir une solution de précision epsilon. Suivant la contrainte, nous gagnons donc un à deux ordres dans la rapidité de convergence par rapport à des approches standard. Finalement, nous faisons quelques expériences numériques qui confirment les résultats théoriques sur de nombreux problèmes. Abstract : This paper presents new algorithms to minimize total variation and more generally l^1-norms under a general convex constraint. The algorithms are based on a recent advance in convex optimization proposed by Yurii Nesterov citeNESTEROV. Depending on the regularity of the data fidelity term, we solve either a primal problem, either a dual problem. First we show that standard first order schemes allow to get solutions of precision epsilon in O(frac1epsilon^2) iterations at worst. For a general convex constraint, we propose a scheme that allows to obtain a solution of precision epsilon in O(frac1epsilon) iterations. For a strongly convex constraint, we solve a dual problem with a scheme that requires O(frac1sqrtepsilon) iterations to get a solution of precision epsilon. Thus, depending on the regularity of the data term, we gain from one to two orders of magnitude in the convergence rates with respect to standard schemes. Finally we perform some numerical experiments which confirm the theoretical results on various problems.

19 - A Three-layer MRF model for Object Motion Detection in Airborne Images. C. Benedek et T. Szirányi et Z. Kato et J. Zerubia. Rapport de Recherche 6208, INRIA, juin 2007. Mots-clés : Aerial images, Change detection, Camera motion, MRF. @TECHREPORT{benedek_INRIARR07, author = {Benedek, C. and Szirányi, T. and Kato, Z. and Zerubia, J.}, title = {A Three-layer MRF model for Object Motion Detection in Airborne Images}, year = {2007}, month = {juin}, institution = {INRIA}, type = {Research Report}, number = {6208}, url = {https://hal.inria.fr/inria-00150805}, pdf = {https://hal.inria.fr/inria-00150805}, keyword = {Aerial images, Change detection, Camera motion, MRF} }

20 - Object extraction using a stochastic birth-and-death dynamics in continuum. X. Descombes et R. Minlos et E. Zhizhina. Rapport de Recherche 6135, INRIA, 2007. Mots-clés : birth and death process, Stochastic modeling, Ondelettes. @TECHREPORT{RR-6135, author = {Descombes, X. and Minlos, R. and Zhizhina, E.}, title = {Object extraction using a stochastic birth-and-death dynamics in continuum}, year = {2007}, institution = {INRIA}, type = {Research Report}, number = {6135}, url = {https://hal.inria.fr/inria-00133726}, pdf = {http://hal.inria.fr/inria-00133726}, keyword = {birth and death process, Stochastic modeling, Ondelettes} } Abstract : We define a new birth and death dynamics dealing with configurations of discs in the plane. We prove the convergence of the continuous process and propose a discrete scheme converging to the continuous case. This framework is developed to address image processing problems consisting in extracting objects. The derived algorithm is applied for tree crown extraction and bird detection from aerial images. The performance of this approach is shown on real data.

haut de la page

navigation : 1 - 2 - 3- ... - 8 - 9 affichage : 10 par page | 20 par page | 50 par page

Ces pages sont générées par