Academic Journal
Geometric models for image processing and understanding
| Τίτλος: | Geometric models for image processing and understanding |
|---|---|
| Συνεισφορές: | Chan, Hei Long (author.), Lui, Lok Ming , 1981- (thesis advisor.), Chinese University of Hong Kong Graduate School. Division of Mathematics. (degree granting institution.) |
| Έτος έκδοσης: | 2017 |
| Συλλογή: | The Chinese University of Hong Kong: CUHK Digital Repository / 香港中文大學數碼典藏 |
| Θεματικοί όροι: | Image processing--Mathematical models, Geometry--Data processing, TA1637 .C63 2017eb |
| Περιγραφή: | M.Phil. ; We consider four problems of image processing and image understanding in this thesis. In this thesis, images are not only referred to the traditional 2D planar rectangular images, but also high dimensional images such as surfaces or even volumetric meshes. ; We divide this thesis into two parts. In the first part, we propose two geometric models for image data processing, so that the given data is well processed and prepared for further analysis. In this part, firstly we address the problem of surface optimization to improve the quality of a given surface. We make use of the Hooke’s law from the elasticity theory of physics and develop an algorithm to greatly improve the triangulation quality of a given surface mesh with minimal distortions to the surface geometry. Also, we deal with 2D image segmentation problem so that the target object can be segmented from a given image data. We introduce a new signature called the Beltrami signature of shapes to capture the geometry of the objects in the image, and hence build a template deformation based model to segment a given image by a domain of user-prescribed topology. ; Given that the input data is now well pre-processed, in the second part of this thesis, we turn our scope to image data understanding and analysis. We propose two geometric models for this purpose. In the first place, we encounter the surface analysis based disease classification problem in which we create an automatic machine to classify the Alzheimer’s disease based on analyzing the newly proposed shape index of the hippocampal surface, using a supervised learning approach. In advance, we deal with the problem of deformation analysis of volumetric meshes to classify between normal and abnormal deformations which are modelled as conformal and non-conformal deformations respectively. To solve the problem, we propose an index called the Anisotrophic Indicator to locally distinguish the deformation, which can be used as a disease diagnosis tool for medical imaging. ... |
| Τύπος εγγράφου: | text |
| Περιγραφή αρχείου: | electronic resource; remote; 1 online resource (vi, 126 leaves) : illustrations (some color); computer; online resource |
| Γλώσσα: | English Chinese |
| Διαθεσιμότητα: | https://julac.hosted.exlibrisgroup.com/primo-explore/search?query=addsrcrid,exact,991039531369503407,AND&tab=default_tab&search_scope=All&vid=CUHK&mode=advanced&lang=en_US https://repository.lib.cuhk.edu.hk/en/item/cuhk-1839297 |
| Rights: | Use of this resource is governed by the terms and conditions of the Creative Commons "Attribution-NonCommercial-NoDerivatives 4.0 International" License (http://creativecommons.org/licenses/by-nc-nd/4.0/) |
| Αριθμός Καταχώρησης: | edsbas.74C9007B |
| Βάση Δεδομένων: | BASE |
| Η περιγραφή δεν είναι διαθέσιμη |