Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Electronic and Information Engineeringen_US
dc.creatorSun, Xin-
dc.identifier.urihttps://theses.lib.polyu.edu.hk/handle/200/5107-
dc.languageEnglishen_US
dc.publisherHong Kong Polytechnic University-
dc.rightsAll rights reserveden_US
dc.titleGA method for image registrationen_US
dcterms.abstractImage registration is one of the elementary steps in image processing. The purpose of image registration is to overlay two images (or more images) of the same scene that are obtained at different times, from different viewpoints, or by different sensors. Typically, image registration can be applied hi medical imaging, remote sensing and computer vision. The images used in the process of registration are called the reference image and the study image. The reference image is the target image while the study image is the transformed image hi the registration process. During the process of image registration, the transformation method is an important component. In brief, the transformation method transforms the study image to the reference image based on the corresponding relationship of the objects. The judgment criteria can be used to calculate the performance after the registration. For the purpose of achieving satisfied performance of the registered image, there are many mathematical optimization methods used hi the process of the transformation method, such as steepest gradient descent optimization, conjugate gradient optimization and Newton methods, etc. However, most of the gradient optimization methods are easily trapped in the local minimum value. Genetic Algorithms (GA) is one of the optimization technologies in global area. Basically, Genetic Algorithms is an adaptive searching algorithm which is inspired by biological evolution. Generally speaking, Genetic Algorithms optimize the fitness function, an objective function, mainly by three GA operations. The optimization results are generated iteratively until the termination conditions are satisfied. In order to overcome the non-optimal solution drawback, Genetic Algorithms is applied in the dissertation as the optimization method. Compared with other gradient based optimization methods, Genetic Algorithms is a derivative free optimization algorithm where only similarity cost function needs to be assessed during the optimization iterations. In my dissertation, the main contributions focus on using Genetic Algorithms to optimize the process of image registration in the platform of MATLAB software. The study image is transformed followed by B-spline transformation method based on the reference image. The performance of the proposed Genetic Algorithms method is compared with those of the other different optimization methods in the dissertation. The benefits and the limitations of GA image registration are analyzed as well.en_US
dcterms.extentix, 101 leaves : ill. (some col.) ; 30 cm.en_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.issued2009en_US
dcterms.educationalLevelAll Masteren_US
dcterms.educationalLevelM.Sc.en_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertations.en_US
dcterms.LCSHImage processing.en_US
dcterms.LCSHGenetic algorithms.en_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
b23030665.pdfFor All Users (off-campus access for PolyU Staff & Students only)4.21 MBAdobe PDFView/Open


Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/5107