Superresolution Imaging: Models and Algorithms

This dissertation, "Superresolution Imaging: Models and Algorithms" by 游展高, Chin-ko, Yau, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled SUPERRESOLUTION IMAGING: MODELS AND ALGORITHMS Submitted by YAU Chin-Ko for the degree of Doctor of Philosophy at The University of Hong Kong in January 2008 In this thesis, two types of super-resolution imaging models were presented: a linear type model using Laplacian matrix as a regularization, and a non-linear type model using total variation (TV) term as a regularization. In the literatures, two types of imaging systems are usually considered: a near eld imaging and a far eld imaging. In the near eld imaging system, the blur- ring from optical devices is considered, whereas in the far eld imaging system, the blurring from the environmental factors is considered. In this thesis, two new models called \medium eld" models, which generalize both of the above imaging systems, were presented. In these new models, the high-resolution restored im- age was obtained by using a maximum-a-posteriori (MAP) estimation technique with a Gaussian prior. The linearity of this model comes from the use of the Laplacian matrix in the regularization. In the rst medium model, the observed low-resolution images were assumed to be rst blurred by the environmental fac- tors, then downsampled, and nally blurred by the optical devices. In the second medium model, the observed low-resolution images were assumed to be blurred by the environmental factors, then blurred by the optical devices, and nally down- sampled. The observed images were all assumed to contain noises. The periodic boundary condition and the zero boundary condition were considered in the blur- ringmatrix. Byusingasuitabletrigonometrictransformstosimplifythecoecient matrix, the Neumann boundary condition, which gives better results than these boundary conditions, can be also considered in all these four models which are Laplacian-based models.For the TV-based model, a unied super-resolution model was presented. The ideaisbasedonthatmissingpixelsoftheobservedimagescanbefoundiftheyare mappedintothehigh-resolutionimagegridandthetotalvariation(TV)inpainting technique is used to nd them. So roughly speaking, a super-resolution imaging model was combined with a TV inpainting model. Furthermore, an extra miss- ing region such as scratches in observed images was allowed in this model. Several observationmodels, whichhavebeenconsideredbysomeotherresearchers, werein- cludedbythisuniedmodel. Thenon-linearitycomesfromtheTVregularization. Afastalgorithmbasedonthexed-pointiterationsandpreconditioningtechnique with factorized sparse inverse preconditioner (FSIP) was developed to solve the problem. The proposed algorithm is faster than the time marching scheme com- monly used to solve the TV type regularization problem. Furthermore, the TV super-resolution model was modied to obtain a super-resolution image by using a sequence of zoomed images. Numerical results were presented to illustrate the proposed algorithm in di(R)er- ent models. They reveal that high quality of the super-resolution images can be obtained by the proposed models. DOI: 10.5353/th_b3955904 Subjects: Resolution (Optics) Imaging systems Laplacian operator Algorithms