Home > Table of Contents


Proceedings of 2009 International Workshop on Information Security and Application (IWISA 2009)

Qingdao, China, November 21-22, 2009

Editors: Feng Gao and Xijun Zhu

AP Catalog Number: AP-PROC-CS-09CN004

ISBN: 978-952-5726-06-0

Page(s): 413-416

An Improved HT Algorithm based Clustering for Multi-objective Detection

        Rong Fei, Duwu Cui, and Bo Hu

Full text:  PDF


This paper describes a new Hough transform (HT) algorithm for multi-objective detection. The core principle of HT—voting principle, which needs to find the maximal voting rate, makes HT impossible to detect many targets from single image synchronously. An improved HT algorithm based on clustering is presented. When the number of targets in the image is known, all of targets can be detected once by this algorithm. The simulation results of iris location with proposed approach reveal that this approach can detect the needed targets in the image once and has great detection effect.

Index Terms

Hough transform; multi-objective detection, c-means clustering

Copyright @ 2009 ACADEMY PUBLISHER — All rights reserved