Universiti Teknologi Malaysia Institutional Repository

Features extraction based on fuzzy clustering and segmentation onto the motion region for medium field surveillance application

Maliki, Mohamad Nansah and Abu Bakar Al-Attas, Syed Abdul Rahman (2004) Features extraction based on fuzzy clustering and segmentation onto the motion region for medium field surveillance application. In: Proceedings - International Conference on Computer Graphics, Imaging and Visuualization, CGIV 2004. IEEE, USA, pp. 70-75. ISBN 978-076952178-7

[img] PDF
Restricted to Repository staff only

341kB

Official URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumbe...

Abstract

Fuzzy-based clustering algorithm is becoming popular in image processing, especially in image segmentation. Based on the degree of the membership function, each pixel can be grouped into “natural group� of pixels according to the spatial coordinate using their values. In this work we present features extraction based on fuzzy clustering and segmentation onto the motion region for medium field surveillance application. Instead of using blob analysis, we believe that fuzzy based clustering algorithm, can also be used to generate different clusters. Incorporating the motion-based segmentation the complexity of the fuzzy clustering will also be reduced because only the motion region will be processed to the clustering algorithm.

Item Type:Book Section
Additional Information:ISBN 0-7695-2178-9 Proceedings - International Conference on Computer Graphics, Imaging and Visualization, CGIV 2004, 26-29 July 2004, Penang.
Uncontrolled Keywords:fuzzy-based clustering algorithm, image processing, image segmentation
Subjects:Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics
T Technology > TR Photography
Divisions:Electrical Engineering
ID Code:9810
Deposited By: Zalinda Shuratman
Deposited On:29 Mar 2010 04:22
Last Modified:07 Mar 2011 09:07

Repository Staff Only: item control page