الفهرس 
Chapter 1 : Visualization an OverviewOnly 14 pages are availabe for public view
 |  | from | 134 |  |  |
| | | from | 134 | | |
Abstract
Visualization is the process of constructing a visual image in the mind. Visualization as a discipline is the art and science of extracting a lot of information in one vision (as a visual image) without reading a lot of information files by the aid of computer graphics, image processing, computer vision and Computer animation. Visualization has a lot of applications such as medical image diagnosis, geography, natural sciences and ecology”. The aim of this work is to adapt image processing algorithms to work in the visualization pipeline to create visual image and highlight the advantages and disadvantages of each one of them. In chapter one, we present the main concepts of visualization, visualization types, techniques, stages, and their use in medical image diagnosis. We also give the pipeline of visualization process to generate visually understandable images from abstract data. In chapter two, we discuss datasets that we work on. These datasets are medical images that include MRI images, CT images and DICOM images. We focus on DICOM images. Then, we discuss preprocessing as a first step in the pipeline of image analysis for Medical Visualization. This include Region Of Interest (ROI) selection and resampling: ROI to visualize one organ whenever the image that we use may take a large slice containing more than one organ, a resampling step to transform data to an isotropic grid to increase the resolution in ii relevant areas. Due to the motion and breathing processes of the patient, noise reduction filters are also discussed. In chapter three, we discuss the main concepts of segmentation of medical image. We study three main types of segmentation: structure techniques, stochastic techniques and hybrid approaches. We develop a level-set technique, Graph- Searching , and discuss 3D Edge-Detection as structure techniques. For stochastic, we study the thresholding technique. Finally, we introduce region growing algorithm as an example of hybrid approach. Other segmentation approaches for medical visualization are discussed too. In chapter four, we discuss rendering techniques. There are two types of rendering, surface rendering for building the three dimensional surface for dicom images and volume rendering for building the entire region of the organ (whole image). For surface rendering, we present two algorithms, contour tracing algorithm as a classical method from image processing to extract boundaries in a digital image and the marching cube algorithm to get the iso surface. For volume rendering, we discuss ray casting algorithm, shear warping algorithm, splatting algorithm and texture mapping algorithm. Ray casting shoot rays from the eye, one per pixel, and find the closest object blocking the path of that ray. Shear wrap generate different views of an object given as volume data. Splatting projects the voxels onto the image-plan. Texture mapping is a method for adding detail, surface texture, or colour to a surface of shape or polygon. In chapter five, simulation results for Rendering and image processing algorithms by MATLAB, Medical Image Processing and Visualization (MIPAV), and 3D Doctors are presented.