A partir d’un ensemble de données interprétées et issues d’une analyse préalable par un opérateur expert (horizons, failles), l’objectif de la thèse est de proposer une segmentation d’une image sismique sous-jacente en parfaite cohérence avec les lois de la géologie. L’originalité de la démarche consistera à développer des techniques de segmentation d’images sismiques, entre autres basées sur des approches de type contours actifs, contraintes par des données interprétées en supplément de propriétés intrinsèques calculées par des procédés automatiques à partir de la donnée traitée sans nécessiter une quelconque supervision contrairement aux travaux existants. Un deuxième axe consistera à ordonnancer automatiquement les horizons (surfaces) interprétés et analyser finement chaque intervalle (le lieu existant entre deux horizons), en prenant en compte son contenu (amplitude, orientation, etc.). Tout cela aboutissant à la reconstruction du pseudo-temps géologique. / The objective of the thesis is to propose a segmentation of an underlying seismic image in perfect coherence with the results of a preliminary analysis by an expert (horizons, faults). laws of geology. The originality of the approach will be to develop techniques for segmenting seismic images, among others based on active contour type approaches, constrained by data interpreted in addition to intrinsic properties calculated by automatic processes from the data processed without requiring any supervision in contrast to existing work. A second axis will be to automatically schedule the horizons (surfaces) interpreted and to analyze each interval (the place between two horizons) finely, taking into account its content (amplitude, orientation, etc.). All this resulted in the reconstruction of the geological pseudo-time.

Cette thèse de doctorat introduit un modèle et un algorithme six-dimensions pour la reconstruction des orientations cristallines locales dans les matériaux polycristallins. Le modèle s’applique actuellement aux données obtenues avec un rayonnement synchrotron (faisceau parallèle et monochromatique), mais il est également possible d’envisager des extensions aux instruments et sources de laboratoire (polychromatique et divergent). Le travail présenté est principalement une extension de la technique connue sous le nom de “Diffraction Contrast Tomography” (DCT) qui permet la reconstruction de la forme et de l’orientation cristalline des grains dans des matériaux polycristallins (avec certaines restrictions concernant la taille et le nombre total de grains ainsi que la mosaicité intragranulaire). / This Ph.D. thesis is about the development and formalization of a six-dimensional tomography method, for the reconstruction of local orientation in poly-crystalline materials. This method is based on a technique known as diffraction contract tomography (DCT), mainly used in synchrotrons, with a monochromatic and parallel high energy X-ray beam. DCT exists since over a decade now, but it was always employed to analyze undeformed or nearly undeformed materials, described by “grains” with a certain average orientation. Because an orientation can be parametrized by the used of only three num- bers, the local orientation in the grains is modelled by a six-dimensional space X6 = R3 ⊗ O3, that is the outer product between a three-dimensional real- space and another three-dimensional orientation-space. This means that for each point of the real-space, there could be a full three-dimensional orientation- space, which however in practice is restricted to a smaller region of interest called “local orientation-space”. The reconstruction problem is then formulated as a global minimisation prob- lem, where the reconstruction of a single grain is the solution that minimizes a functional. There can be different choices for the functionals to use, and they depend on the type of reconstructions one is looking for, and on the type of a priori knowledge is available. All the functionals used include a data fidelity term which ensures that the reconstruction is consistent with the measured diffraction data, and then an additional regularization term is added, like the l1-norm minimization of the solution vector, that tries to limit the number of orientations per real-space voxel, or a Total Variation operator over the sum of the orientation part of the six-dimensional voxels, in order to enforce the homogeneity of the grain volume. When first published, the results on synthetic data from the third chapter high- lighted some key features of the proposed framework, and showed that it was in principle possible to extend DCT to the reconstruction of moderately de- formed materials, but it was unclear whether it could work in practice. The following chapters instead confirm that the proposed framework is viable for reconstructing moderately deformed materials, and that in conjunction with other techniques, it could also overcome the limitations imposed by the grain indexing, and be applied to more challenging textured materials.