ISU Electrical and Computer Engineering Archives

Medium resolution Computed Tomography through phosphor screen detector and 3D image analysis

Sheikh, Nadia (2006) Medium resolution Computed Tomography through phosphor screen detector and 3D image analysis. Masters thesis, Iowa State University.

Full text available as:

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

Abstract

Computed Tomography (CT) technology allows the cross sectional imaging of objects nondestructively and is of great importance because of its capability of detecting interior defects. CT technology is especially widely used in the areas of material industries, airspace industries and medical diagnostics. However this detail of information comes at a price. Modern CT data acquisition systems collect huge amounts of data making the processing of that data a significant challenge. In order to deal with continuing demand of large object 3D CT technology and increased 3D CT data size we have developed a large field of view 3D CT scan capability addressing the required resolution (100 μ) and handling the data acquisition, transfer, processing, viewing and information extraction. The new phosphor screen detector employing CCD camera with approximately 44cm x 30cm active imaging area and 3073 x 2048 pixel size, scans large objects at medium resolution. Phosphor screen computed tomography system control software developed in this thesis integrates image acquisition, image calibration, motion control and CT Scan. CT data generated as a result of a scan requires a 3D visualization tool to fulfill goal of inspection and analysis of object. The 3D visualization tool designed at Center of Non Destructive Evaluation employs ray cast volume rendering method to visualize volume CT data on PC. While the ability to get an overview of these large 3D CT data sets (3GB) is powerful, the lack of tools to extract information from this qualitative display limits the 3D CT capability. This research has added another tool to the existing software that will be used to identify and separate features in a sample based on identifying neighbor approach. Algorithms for image analysis and processing like percentage porosity analysis and correction of rogue/bad pixels are also part of the research. The result in dealing with the data acquisition and data manipulation issues is a significant extension of 3D CT imaging capabilities.

EPrint Type:Thesis (Masters)
Subjects:Electrical Engineering > ELECTROMAGNETICS & NONDESTRUCTIVE EVALUATION > Computer and Communication Networking
ID Code:234
Identification Number:TR-2006-04-14
Deposited By:Nadia Sheikh
Deposited On:17 April 2006

Archive Staff Only: edit this record