Rozpoznávání objektů na snímcích v medicíně

Abstract

The goal of this diploma thesis is a proposal and implementation of module which will be helpful for a user for an advanced recognition of an object of interest, e.g. a profile of carotid artery or a brainstem contained in medical pictures, or other pictures. Modules enable initial adjustment with help of wide range of different image filters which help to achieve more precise work of the second module – the automatic segmentation itself. The automatic segmentation will be provided by methods of active contours using the method Gradient Vector Flow and Fast Marching Level Sets. Both modules will be implemented in JAVA with JAI framework and covered to the FOTOM NG.