Image segmentation by scalable spatial information

Pao Yue-kong Library Electronic Theses Database

Image segmentation by scalable spatial information

 

Author: Ho, Kam-wai Raymond
Title: Image segmentation by scalable spatial information
Degree: M.Sc.
Year: 2000
Subject: Image processing
Hong Kong Polytechnic University -- Dissertations
Department: Multi-disciplinary Studies
Dept. of Electronic and Information Engineering
Pages: 129 leaves : ill. ; 30 cm
Language: English
InnoPac Record: http://library.polyu.edu.hk/record=b1517707
URI: http://theses.lib.polyu.edu.hk/handle/200/4370
Abstract: A maximum a posteriori spatial probability (MASP) segmentation algorithm that performs pixel-by-pixel segmentation by utilizing spatial information which is the combined properties of the segmented pixel and its neighbour pixels. These properties are derived from the geographically locations and the number of neighbourhood pixels being utilized and they can influence the segmentation performance. In this project, simulations of segmenting synthetic images are processed to illustrate the segmentation results by using different locations of neighbourhood pixels and the accuracy improvement while using more neighbourhood pixels. A wide variety of neighbourhood configurations, make up of from one-neighbour neighbourhood scheme to eight-neighbour neighbourhood scheme, have been implemented to investigate the accuracy and processing speed. In the simulation of different locations of neighbourhood pixels, it shows that segmentation accuracy is independent to the locations of neighbourhood pixels. In the simulation of different numbers of neighbourhood pixels, it shows that when the number of neighbour pixels increase, the segmentation accuracy increases but the processing speed decreases. Compared the segmentation performance in the simulations, a 4-neighbour neighbourhood scheme is taken as optimal neighbourhood configuration.

Files in this item

Files Size Format
b15177075.pdf 4.679Mb PDF
Copyright Undertaking
As a bona fide Library user, I declare that:
  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.
By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

     

Quick Search

Browse

More Information