الفهرس 
Chapter1 INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 202 |  |  |
| | | from | 202 | | |
Abstract
Image segmentation is an essential step in many advanced imaging applications, e.g., object tracking, pattern recognition, volume measurements, medical image analysis, and 3D rendering. Accurate image segmentation is required in most of the medical imaging applications; for volume measurements, medical diagnosis, and in the image guided procedures. Medical image analysis is a fast growing research area in the image processing community due to the importance of its application in the field of medicine.
Among the several types of images, magnetic resonance images (MRI), which represent the intensity variation of radio waves generated by biological systems when exposed to radio frequency pulses, have proved to be the most effective imaging modality for imaging the inner tissues of the human. For that, the image processing community has extensively developing algorithms for analyzing those images, in order to achieve accurate segmentation that may facilitate the detection of various pathological conditions affecting brain parenchyma, radiotherapy treatment and planning, surgical planning and simulations, and three-dimensional (3D) visualization of brain matter for diagnosis and abnormality detection. The principal goal of the work in this thesis is to produce an offline system for the accurate segmentation of the medical images.
In this thesis, we have developed two new multiresolution algorithms for image segmentation that extend the well-known Expectation Maximization (EM) algorithm. These are, the Gaussian Multiresolution EM algorithm, GMEM, and the Wavelet Multiresolution EM algorithm,