City University of Hong Kong
DSpace
 

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Manufacturing Engineering and Engineering Management  >
MEEM - Master of Philosophy  >

Please use this identifier to cite or link to this item: http://hdl.handle.net/2031/4525

Title: A series of surface fitting strategies for procedure optimization in reconstructing high quality surfaces
Other Titles: Yi xi lie pei he fang an di zao geng jia zhi chong zu gao zhi qu mian cheng xu
一系列配合方案締造更佳之重組高質曲面程序
Authors: So, Yin Ching (蘇燕清)
Department: Dept. of Manufacturing Engineering and Engineering Management
Degree: Master of Philosophy
Issue Date: 2005
Publisher: City University of Hong Kong
Subjects: Computer-aided design
Surfaces, Algebraic -- Data processing
Notes: CityU Call Number: T385.S629 2005
Includes bibliographical references (leaves 123-125)
Thesis (M.Phil.)--City University of Hong Kong, 2005
ix, 158 leaves : ill. (some col.) ; 30 cm.
Type: Thesis
Abstract: Using surface fitting algorithms to reconstruct geometrical features of 3D objects is used intensively for a wide range of applications from toy production to medical surgery. However, the procedures are always prone to intensive labour involvements and require a lot of technical skills. Recently, academic researches push forward the study on improving surface fitting results to achieve a better quality with simple processing procedures. Although using surface fitting techniques has been diversified into both engineering and non-engineering applications, complicated works are largely needed on educating technical skills and finding better ways to produce quality surfaces in the present of computer-aid modelling tools. The research aims at developing strategies that enhance the processing procedures in surface fitting for achieving a required quality at the shortest time. B-spline surface modelling algorithms are employed as the fitting tools. In fitting a B-spline surface from a group of point clouds, some essential modelling procedures are needed. Two major procedures are the selection of surface generating methods and the determination of fitting parameters. In fact, some developed surface generating methods are available for different shapes’ surfaces. A shorter surface fitting processing time can be achieved when appropriate generating methods are chosen, and a better quality surface is produced from appropriately assigning the fitting parameters. The main objectives of the research are to categorize appropriate surface generating methods for various shapes’ surfaces, and develop a mathematical tool to assist the assignment of an important fitting parameter, i.e. the number of control points. One of the common modelling software, Surfacer was chosen as the surface-fitting tool for the entire investigation. In studying the selection of surface generating methods, some guidelines were developed for quickly choosing the appropriate methods to fit a surface. In the study of assisting parameter assignments, fuzzy logic was adopted to develop the algorithm on assigning the number of control points. Fit Free-form and Lofting strategies as supported by Surfacer were studied intensively. Several case studies on the fitting of human face profiles were deeply investigated, and the proposed algorithm on assigning the control point number was evaluated. Using the proposed fitting parameter assignment’s strategies can speed up the entire processing time in surface reconstructions, and achieve the required quality of surface fitting with minimum human involvements. Successfully applying the proposed strategies in surface fitting, the traditional intensive learning on surface modelling tools can be simplified, and also confidence for unskilled designers can be built up in using sophisticated computer-aided surface modelling tools.
Online Catalog Link: http://lib.cityu.edu.hk/record=b1988594
Appears in Collections:MEEM - Master of Philosophy

Files in This Item:

File Description SizeFormat
fulltext.html161 BHTMLView/Open
abstract.html161 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer