Th. doct.--Signal et images--Paris--ENST, 1995. / Textes en français ou en anglais. Bibliogr. p. 167-171. Résumé en français et en anglais.

Th.--Signal et images--Paris--ENST, 1995. / En apendice, choix d'articles en anglais. Bibliogr. p. 73-79. Ouvrage porte par erreur : ISSN 1242-5125.

Th. doct.--Signal et images--Paris--ENST, 1994. / En appendice, choix de textes en anglais et en français extr. de diverses publications. Bibliogr., 4 p.

Th. doct.--Signal et images--Paris--ENST, 1993. / Notes bibliogr.

Th.--Sci. phys.--Nancy--I.N.P.L., 1980.

This thesis studies statistical image processing of high frequency ultrasound imaging, with application to in-vivo exploration of human skin and noninvasive lesion assessment. More precisely, Bayesian methods are considered in order to perform tissue segmentation in ultrasound images of skin. It is established that ultrasound signals backscattered from skin tissues converge to a complex Levy Flight random process with non-Gaussian _-stable statistics. The envelope signal follows a generalized (heavy-tailed) Rayleigh distribution. Based on these results, it is proposed to model the distribution of multiple-tissue ultrasound images as a spatially coherent finite mixture of heavy-tailed Rayleigh distributions. Spatial coherence inherent to biological tissues is modeled by a Potts Markov random field. An original Bayesian algorithm combined with a Markov chain Monte Carlo method is then proposed to jointly estimate the mixture parameters and a label-vector associating each voxel to a tissue. The proposed method is successfully applied to the segmentation of in-vivo skin tumors in high frequency 2D and 3D ultrasound images. This method is subsequently extended by including the estimation of the Potts regularization parameter B within the Markov chain Monte Carlo (MCMC) algorithm. Standard MCMC methods cannot be applied to this problem because the likelihood of B is intractable. This difficulty is addressed by using a likelihood-free Metropolis-Hastings algorithm based on the sufficient statistic of the Potts model. The resulting unsupervised segmentation method is successfully applied to tridimensional ultrasound images. Finally, the problem of computing the Cramer-Rao bound (CRB) of B is studied. The CRB depends on the derivatives of the intractable normalizing constant of the Potts model. This is resolved by proposing an original Monte Carlo algorithm, which is successfully applied to compute the CRB of the Ising and Potts models.

Thèse de doctorat--Signal et images--Paris--ENST, 2006. / En appendice, un article en anglais. Bibliogr. p. 147-151. Résumé en français et en anglais.

Thèse de doctorat : Images et Systèmes : Villeurbanne, INSA : 2002. / Titre provenant de l'écran-titre. Bibliogr. p.140.

Thèse de doctorat : Sciences de l'ingénieur : Paris Est : 2008. / Titre provenant de l'écran-titre. Bibliogr. : 150 réf.

Thèse de doctorat : Informatique : Ecully, Ecole centrale de Lyon : 2008. / Titre provenant de l'écran-titre. 162 références.