| M. Styner. Combined Boundary-Medial Shape Description of Variable Biological Objects. PhD thesis, University of North Carolina at Chapel Hill, Chapel Hill, NC, 2001. |
....of feature selection paradigm to the shape characterization problem have received less attention. However, there has been considerable work in the literature in using classification techniques to detect, localize and describe the anatomical di#erences in the shape between di#erent populations [11, 10, 27, 8, 18, 19, 32]. 3 Methods This section contains the details of the new feature selection method for shape classification, which is an extension of an existing feature selection algorithm by Bradley and Mangasarian [3] The novelty of our method is that it searches for an optimal set of windows of features, as ....
M. Styner. Combined Boundary-Medial Shape Description of Variable Biological Objects. PhD thesis, University of North Carolina at Chapel Hill, Chapel Hill, NC, 2001.
.... because natural operations such as bending, widen 12 ing and elongation are easily implemented [17] 34] 36] Styner automatically computed m rep templates with fixed branching topology that through deformation accurately fit populations of hippocampi, amygdalae and other subcortical organs [32]. In this paper we show that the continuous medial representation can be applied to segment anatomical objects in medical images, following the same deformable templates framework used for discrete m reps in work of Joshi, et al. [17] As an example, we automatically segment a vertebral image from ....
M. Styner. Combined Boundary-Medial Shape Description of Variable Biological Objects. PhD thesis, University of North Carolina at Chapel Hill, Chapel Hill, NC, 2001.
.... similar objects, how can we automatically compute a stable medial model in the presence of shape variability The following sections describe the our scheme that construct a medial m rep model from an object population described by boundary parameterization using spherical harmonics (SPHARM) see [13] for more details) In overview, our scheme is divided into 3 steps and visualized in Fig. 1. We first define a shape space using Principal Component Analysis. From this shape space we generate the medial model in two steps. First we compute the common branching topology. Then we calculate the ....
Martin Styner, Combined Boundary-Medial Shape Description of Variable Biological Objects, Ph.D. thesis, UNC Chapel Hill, Computer Science, June 2001, available at www.ia.unc.edu/public/styner/docs/diss.html.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC