The work embodied in this thesis describes software techniques developed to analyse digital inline holograms of suspended particle fields, particularly in aquatic environ- ments. The primary motivation behind this work has been development of tech- niques to extract useable information from individual holograms within holovideos, producing focused silhouettes of recorded plankton and other particulates with min- imal user intervention. Two automated focusing algorithms are developed and presented in this work, both of which obtain comparable results for holograms of sparse plankton populations. The first approach is based on rectangular regions of interest (ROIs), which are aligned to (x, y) dimensions, and localise particles within the two-dimensional recon- structed planes obtained from holovideo frames. Due to poor immunity to particle merging when applied to denser particle fields, a second approach was developed using arbitrary polygons with which to localise particle positions in reconstructed planes. This new approach offers a greater immunity to the merging of particles lying in close proximity in the (x, y) dimensions of the hologram, and allows better particle localisation for high density particle holograms. Both ROI and polygon based particle localisation are explored to identify strengths and weaknesses, and complete automated scanning procedures developed in both cases. Examples are provided of typical output from automated scanning algorithms when applied to a number of sample holograms, and areas of weakness highlighted for future work.

No description available.

The clinical success of Total Hip Arthroplasty is enhanced by matching hip implant geometry to femoral geometry. Traditionally, the shapes of hip implant designs have been based on data collected from patient populations using X-ray, CT Scan, digitized sliced bone, and other physical methods. The morphology of interest and the frame of reference often vary across researchers and the resulting numeric data are difficult to use in a Computer Aided Design package to build an implant model. This goal of this thesis was to develop procedures and automated programs for the design and evaluation of femoral hip implants using CT data. The procedures bridge the gap between patient specific "custom" prosthesis design and the so-called "averaged femur" implant designs. By automating the measurement, orientation and averaging of any user selected grouping of femora, these programs allow construction of an average femur template for small subsets of a population or populations. The programs allow grouping of femora using either dimensional attributes and/or patient attributes such as pathology, ethnic background, etc. The average femur template created for each group can then be used as the design boundary for a discrete implant size. Additional functionality is also provided for comparing average femur templates to their constitutive femora and for comparison of average femur templates against each other. To illustrate that the goals of this theses were realized, an example of the use of the system for a population of 192 Japanese is included in this thesis. The criteria for evaluation of the average femur templates was the gap or interference of the template to that of each constitutive femur along the medial and lateral endosteal contours. Testing 24 template sizes, the average of the worst fitting template to femur combination for each size resulted in a gap or interference of less than 2mm in the metaphysis and less than 4mm proximally.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

This thesis presents an efficient computational voxelization approach that utilizes the graphics pipeline. Our approach is hybrid in that it performs a precise gap-free computational voxelization, employs fixed-function components of the GPU, and utilizes the stages of the graphics pipeline to improve parallelism. This approach makes use of the latest features of OpenGL and fully supports both conservative and thin voxelization. In contrast to other computational voxelization approaches, this approach is implemented entirely in OpenGL, and achieves both triangle and fragment parallelism through its use of both the geometry and fragment shaders. A novel approach utilizing the graphics pipeline to complement geometric triangle intersection computations is presented. By exploiting features of the existing graphics pipeline we are able to rapidly compute accurate scene voxelization in a manner that integrates well with existing OpenGL applications, is robust across many different models, and eschews the need for complex work/load-balancing schemes. / Graduation date: 2013

Thesis (M.S.)--University of Florida, 2001. / Title from first page of PDF file. Document formatted into pages; contains vii, 45 p.; also contains graphics. Vita. Includes bibliographical references (p. 43-44).

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

Thesis (M.Sc.)--York University, 2006. Graduate Programme in Computer Science. / Typescript. Includes bibliographical references (leaves 157-162). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR29596