Abstract
lmage segmentation is a fundamental first step image processing
technique which helps attaining the objectives of computer vision.
The primary objective of this thesis is to enhance existing image
segmentation methods based on Monte Carlo integration.
We have suggested Monte Carlo techniques to find the ensemble
average of neighboring image properties. This way, more precise
image operations such as edge detection, merging areas of similar
properties and meshing techniques will be possible.
Examples from everyday's activities using advanced computer
graphics and image analysis techniques are used. Future work
includes using of enhanced algorithms in medical image
segmentation and analysis.
