Volumetric image analysis: optical flow, registration and segmentation

by Ruwan Bandara Tennakoon

Institution: Swinburne University of Technology
Department: Faculty of Science, Engineering and Technology
Degree: PhD
Year: 2015
Keywords: Volumetric images; Image registration; Robust model fitting; 3D image analysis
Record ID: 1032343
Full text PDF: http://hdl.handle.net/1959.3/393096


This thesis presents more accurate and efficient methods for volumetric image analysis in terms of Optical Flow, Registration and Segmentation. Firstly, a relationship between the estimation accuracy and the required amount of smoothness for motion estimation from a robust statistics perspective is developed. Next, a fast and accurate non-rigid registration method for intra-modality volumetric images that exploits the information provided by an order statistics-based segmentation method, to find the important regions for registration is presented. Finally, two new methods that improve the accuracy and efficiency of the identification of underlying structures in data that is contaminated with noise and outliers are proposed.