Fast Visual Search by Global Pruning and Spiral Scanning

Shingo TAGUCHI  Shinichiro OMACHI  Hirotomo ASO 

Publication
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)  Vol.J90-D  No.7  pp.1765-1772
Publication Date: 2007/07/01
Online ISSN: 1881-0225
Print ISSN: 1880-4535
Type of Manuscript: PAPER
Category: 
Keyword: 
object detection,  color histogram,  active search, 

Full Text(in Japanese): PDF(365.8KB)

Summary: 
Object detection in an image has a lot of useful applications. It has been reported that color information has advantages in detecting objects. The active search introducing adaptive pruning is known as a method that obtains theoretically the same solution as the exhaustive search. However, the scanning rule is not defined explicitly in the active search. In this paper, a high-speed visual search algorithm that adopts search strategy by defining scanning rules is proposed. Experimental results show that the proposed method outperforms the conventional method.