University of Strathclyde
Nonlinear non-Gaussian algorithms for signal and image processing
Abstract
dc:description.abstractThis thesis is initially concerned with solving the Blind Source Separation (BSS)problem. The BSS problem has been found to occur frequently in problemsexisting in various Scientific and Engineering application areas. The basic ideaof the BSS problem is to separate a collection of mixed data into its underlyinginformation components.To tackle the BSS problem two related methodologies have been utilized extensively throughout the literature. The first approach is by utilizing the statisticaltechnique Independent Component Analysis (ICA). This method utilizes a transformation that maximizes the statistical independence of the mixed data components. The second approach is based on the Approximate Joint Diagonalization(AJD) of a set of target matrices, either the time delayed correlation matrices ormatrix slices of the fourth order cumulant tensor. This approximate diagonalization results in matrices which are maximally diagonal. Within this thesis bothof the above approaches are utilized within an adaptive gradient descent settingto tackle the BSS problem.The first contribution within this thesis is the novel application of the MatrixMomentum optimization framework to perform ICA, via the optimization of aMutual Information based cost function. The algorithm is shown to give Newtonlike performance with low computational cost.The second contribution within this thesis is the first application of the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm to jointly diagonalize a set of time delayed correlation matrices.As a result of the above work it was also found that the SPSA algorithm couldalso be applied to the problem of Image Registration. Currently one of themost popular methods of solving the Image Registration problem is based on the maximization of the Mutual Information between the images. The final contribution within this thesis is the application of the SPSA algorithm to othernovel Information Theoretic cost functions to perform Image Registration.
Degree
thesis:*- Name dc:type.qualificationname
- phd
- Level dc:type.qualificationlevel
- doctoral-pg
- Grantor dc:publisher.institution
- University of Strathclyde
- Year dc:date.issued
- 2007
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Morison, Gordon
- Advisor dc:contributor.advisor
-
- Durrani, Tariq S.
Rights
- Licence dc:rights.uri
Identifiers
dc:identifier.*- Identifier
-
uk.bl.ethos.501874
T12169