Beta 1


Title Omni-Directional Image Deblurring
Author Tigkos, Konstantinos
Supervisor Hansen, Per Christian (Scientific Computing, Institut for Informatik og Matematisk Modellering, Danmarks Tekniske Universitet, DTU, DK-2800 Kgs. Lyngby, Denmark)
Aanæs, Henrik (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Institution Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
Thesis level Master's thesis
Year 2011
Abstract This thesis deals with deblurring images corrupted by rotational motion around any possible axis. The study involves developing adequate forward models being able to accurately describe this type of degradation, investigating reconstruction algorithms and finally developing methods that are able to counter this particular problem. To that end, relevant theory is presented and a rotational motion blur simulator is developed in order to compare the forward models proposed. A study about reconstruction algorithms was also carried out in order to make a decision about the methods on which the solution of the particular problem would be based upon. The underlying algebraic properties of this problem are also investigated and the stage is set for extensive reconstruction experiments. The reconstruction experiments, apart from the usual behavior of the reconstruction methods, also revealed some interesting facts about artifacts occurring when deblurring images degraded by motion blur. Some solutions were proposed and a brief investigation into stopping rules for the reconstruction methods was carried out. Most of the code produced throughout this study was included in the MATLAB software package developed.
Imprint Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Series IMM-M.Sc.-2011-11
Fulltext
Original PDF ep11_11.pdf (4.07 MB)
Admin Creation date: 2011-03-23    Update date: 2011-03-23    Source: dtu    ID: 275655    Original MXD