Defining and computing stable representations of volume shapes from discrete trace using volume primitives: Application to 3D image analysis in soil science

This paper presents an innovative approach for defining and computing stable (intrinsic) representations describing volume shapes from discrete traces without any a priori information. We assume that the discrete trace of the volume shape is defined by a binary 3D image where all marked points define the shape. Our basic idea is to describe the corresponding volume using a set of patches of volume primitives (bowls, cylinders, cones...). The volume primitives representation is assumed to optimize a criterion ensuring its stability and including a characterization of its scale (trade-off: fitting errors/number of patches). Our criterion takes also into account the preservation of topological properties of the initial shape representation (number of connected components, adjacency relationships...). We propose an efficient computing way to optimize this criterion using optimal region growing in an adjacency valuated graph representing the primitives and their adjacency relationships. Our method is applied to the modelling of porous media from 3D soil images. This new geometrical and topological representation of the pore network can be used to characterize soil properties.

[1]  James F. Blinn,et al.  A generalization of algebraic surface drawing , 1982, SIGGRAPH.

[2]  Guillaume Caumon,et al.  Visualization of grids conforming to geological structures: a topological approach , 2005, Comput. Geosci..

[3]  J. Mallet,et al.  Building and Editing a Sealed Geological Model , 2004 .

[4]  Pierre Hellier,et al.  Cooperation between level set techniques and dense 3D registration for the segmentation of brain structures , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[5]  Ivan Bricault,et al.  From Volume Medical Images to Quadratic Surface Patches , 1997, Comput. Vis. Image Underst..

[6]  D. Mumford,et al.  Optimal approximations by piecewise smooth functions and associated variational problems , 1989 .

[7]  E. J. Morton,et al.  Non-invasive imaging of roots with high resolution X-ray micro-tomography , 2003, Plant and Soil.

[8]  Nicholas Ayache,et al.  Epidaure: A research project in medical image analysis, simulation, and robotics at INRIA , 2003, IEEE Transactions on Medical Imaging.

[9]  Pascal Frey,et al.  MEDIT : An interactive Mesh visualization Software , 2001 .

[10]  Olivier Monga,et al.  An Optimal Region Growing Algorithm for Image Segmentation , 1987, Int. J. Pattern Recognit. Artif. Intell..

[11]  Hans-Jörg Vogel,et al.  Quantitative morphology and network representation of soil pore structure , 2001 .

[12]  Dimitris N. Metaxas,et al.  Dynamic 3D models with local and global deformations: deformable superquadrics , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[13]  Edith Perrier,et al.  DXSoil, a library for 3D image analysis in soil science , 2002 .

[14]  William E. Lorensen,et al.  Marching cubes: A high resolution 3D surface construction algorithm , 1987, SIGGRAPH.

[15]  Jean Francois Delerue.  Segmentation 3d, application a l'extraction de reseaux de pores et a la caracterisation hydrodynamique des sols , 2001 .

[16]  Olivier D. Faugeras,et al.  Polyhedral approximation of 3-D objects without holes , 1984, Comput. Vis. Graph. Image Process..

[17]  Baba C. Vemuri,et al.  Shape Modeling with Front Propagation: A Level Set Approach , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[18]  G. Pérès.  Identification et quantification in situ des interactions entre la diversité lombricienne et la macro-bioporosité dans le contexte polyculture breton. , 2003 .