Date of Award
6-13-2014
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Department
Applied Science
First Advisor
Mariofanna Milanova
Abstract
Natural images, in the context of this work, are images that contain complicated, local covariance structures in their statistical analysis. Owing to compound textural features, intensity inhomogeneity, image layers and variations of statistics inherent in such images, segmenting complicated images into areas of similarity is a challenging task. In this work, a morphological active contour is developed to increase efficiency of current active contour schemes and a fuzzy clustering energy is incorporated into the active contour algorithm to increase accuracy and flexibility. The savings in computational efficiency garnered from using a morphological curve evolution rather than a partial differential equation and corresponding Euler-Lagrange equations combined with the expert knowledge garnered from a fuzzy logic scheme translates into a highly accurate and efficient segmentation method for complex images.
Recommended Citation
Fox, Victoria Lynn, "Morphological Active Contours for Texture and Multiphase Segmentation" (2014). Theses and Dissertations. 508.
https://research.ualr.edu/etd/508
