Filter-based reconstruction methods for tomography

Leiden Repository

Filter-based reconstruction methods for tomography

Type: Doctoral Thesis
Title: Filter-based reconstruction methods for tomography
Author: Pelt, D.M.
Issue Date: 2016-05-03
Keywords: Tomography
Image reconstruction
Inverse problems
Approximation
Filtered backprojection
Machine learning
Abstract: In X-ray tomography, a three-dimensional image of the interior of an object is computed from multiple X-ray images, acquired over a range of angles. Two types of methods are commonly used to compute such an image: analytical methods and iterative methods. Analytical methods are computationally efficient, but in many applications, they produce reconstructions that are not accurate enough for further analysis. More accurate reconstructions can be obtained by using (regularized) iterative methods, but these can have computational costs that are too high to be used in practice. In this thesis, new reconstruction methods are developed that combine the analytical and algebraic approaches, resulting in methods that are as computationally efficient as analytical methods, but with a reconstruction accuracy of iterative methods. Analytical methods allow for changing their filter without increasing the needed computation time. We use this freedom in filter choice to develop new filter-based reconstruction methods, which are based on the analytical FBP method with specific filters. The filters can be defined and computed in different ways, and can depend on the acquisition geometry, the scanned object, and/or a separate pre-computing step. Several filter-based methods are introduced in this thesis and reconstruction results are compared with other popular methods.
Promotor: Supervisor: K.J. Batenburg
Faculty: Science
University: Leiden
Uri: urn:isbn:9789462332690
Handle: http://hdl.handle.net/1887/39638
 

Files in this item

Description Size View
application/pdf Full Text 48.94Mb View/Open
application/pdf Cover 440.3Kb View/Open
application/pdf Title Pages_Contents 180.7Kb View/Open
application/pdf Chapter 1 4.816Mb View/Open
application/pdf Chapter 2 21.14Mb View/Open
application/pdf Chapter 3 603.9Kb View/Open
application/pdf Chapter 4 3.592Mb View/Open
application/pdf Chapter 5 5.753Mb View/Open
application/pdf Chapter 6 4.575Mb View/Open
application/pdf Chapter 7 10.12Mb View/Open
application/pdf Bibliography_Publications 223.0Kb View/Open
application/pdf Summary in Dutch 2.413Mb View/Open
application/pdf Curriculum Vitae_Acknowledgements 123.5Kb View/Open
application/pdf Propositions 115.3Kb View/Open

This item appears in the following Collection(s)