Catálogo de publicaciones - libros

Motivated by the problem of analyzing shapes of fiber tracts in DT-MRI data, we present a geometric framework for studying shapes of open curves in ℝ. We start with a space of unit-length curves and define the shape space to be its quotient space modulo rotation and re-parametrization groups. Thus, the resulting shape analysis is invariant to parameterizations of curves. Furthermore, a Riemannian structure on this quotient shape space allows us to compute geodesic paths between given curves and helps develop algorithms for: (i) computing statistical summaries of a collection of curves using means and covariances, and (ii) clustering a given set of curves into clusters of similar shapes. Examples using fiber tracts, extracted as parameterized curves from DT-MRI images, are presented to demonstrate this framework.

In the area of quality control by vision, the reconstruction of 3D curves is a convenient tool to detect and quantify possible anomalies. Whereas other methods exist that allow us to describe surface elements, the contour approach will prove to be useful to reconstruct the object close to discontinuities, such as holes or edges.

We present an algorithm for the reconstruction of 3D parametric curves, based on a fixed complexity model, embedded in an iterative framework of control point insertion. The successive increase of degrees of freedom provides for a good precision while avoiding to over-parameterize the model. The curve is reconstructed by adapting the projections of a 3D NURBS snake to the observed curves in a multi-view setting. The optimization of the curve is performed with respect to the control points using an gradient-based energy minimization method, whereas the insertion procedure relies on the computation of the distance from the curve to the image edges.

This study extended the computation of GLCM (gray level co-occurrence matrix) to a three-dimensional form. The objective was to treat hyperspectral image cubes as volumetric data sets and use the developed 3D GLCM computation algorithm to extract discriminant volumetric texture features for classification. As the kernel size of the moving box is the most important factor for the computation of GLCM-based texture descriptors, a three-dimensional semi-variance analysis algorithm was also developed to determine appropriate moving box sizes for 3D computation of GLCM from different data sets. The developed algorithms were applied to a series of classifications of two remote sensing hyperspectral image cubes and comparing their performance with conventional GLCM textural classifications. Evaluations of the classification results indicated that the developed semi-variance analysis was effective in determining the best kernel size for computing GLCM. It was also demonstrated that textures derived from 3D computation of GLCM produced better classification results than 2D textures.

In this work, we introduce a robust energy model for multiview 3D reconstruction that fuses silhouette- and stereo-based image information. It allows to cope with significant amounts of noise without manual pre-segmentation of the input images. Moreover, we suggest a method that can globally optimize this energy up to the visibility constraint. While similar global optimization has been presented in the discrete context in form of the maxflow-mincut framework, we suggest the use of a continuous counterpart. In contrast to graph cut methods, discretizations of the continuous optimization technique are consistent and independent of the choice of the grid connectivity. Our experiments demonstrate that this leads to visible improvements. Moreover, memory requirements are reduced, allowing for global reconstructions at higher resolutions.

In this paper we present and evaluate a new Bayesian method for range image segmentation. The method proceeds in two stages. First, an initial segmentation is produced by a randomized region growing technique. The produced segmentation is considered as a degraded version of the ideal segmentation, which should be then refined. In the second stage, pixels not labeled in the first stage are labeled by using a Bayesian estimation based on some prior assumptions on the regions of the image. The image priors are modeled by a new Markov Random Field (MRF) model. Contrary to most of the authors in range image segmentation, who use only surface smoothness MRF models, our MRF takes into account also the smoothness of region boundaries. Tests performed with real images from the ABW database show a good potential of the proposed method for significantly improving the segmentation results.

This paper presents a two-step algorithm to perform automatic extraction of vessel tree on angiogram. Firstly, the approximate vessel centerline is modeled as marked point process with each point denoting a line segment. A prior model is proposed to incorporate the geometrical and topological constraints of segments through potentials on the interaction and the type of segments. Data likelihood allows for the vesselness of the points which the segment covers, which is computed through the Hessian matrix of the image convolved with 2-D Gaussian filter at multiple scales. Optimization is realized by simulated annealing scheme using a Reversible Jump Markov Chain Monte Carlo (RJMCMC) algorithm. Secondly, the extracted approximate vessel centerline, containing global geometry shape as well as location information of vessel, is used as important guide to explore the accurate vessel edges by combination with local gradient information of angiogram. This is implemented by morphological homotopy modification and watershed transform on the original gradient image. Experimental results of clinical digitized coronary angiogram are reported.

Surface reconstruction from implicit data of sub-randomly distributed 3D points is the key work of extracting explicit information from LiDAR data. This paper proposes an approach of extended snake theory to surface reconstruction from LiDAR data. The proposed algorithm approximates a surface with connected planar patches. Growing from an initial seed point, a surface is reconstructed by attaching new adjacent planar patches based on the concept of minimizing the deformable energy. A least-squares solution is sought to keep a local balance of the internal and external forces, which are inertial forces maintaining the flatness of a surface and pulls of observed LiDAR points bending the growing surface toward observations. Experiments with some test data acquired with a ground-based LiDAR demonstrate the feasibility of the proposed algorithm. The effects of parameter settings on the delivered results are also investigated.