Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/26940
Title: Object popping-out and characterization based on the human visual mechanism
Authors: Fu, H
Chi, Z 
Feng, D
Keywords: Human’s attention model
Object detection
Image representation
Issue Date: 2006
Source: International journal of information technology, 2006, v. 12, no. 5, p. 56-64 How to cite?
Journal: International journal of information technology 
Abstract: Semantic image understanding is the basis of a well-performed image management system on a large database whereas current image representations are neither powerful nor semantic-based. In this paper, the human vision mechanism from psychological studies is employed in order to understand an image at a higher level by constructing a layered representation structure based on the image segmentation result. An iterative object popping-out algorithm is proposed to locate the visually attentive objects from a downsized image by maximizing a global attention function designed according to the human’s attention model. The extracted objects are then characterized by using the original image while leaving the background as a rough description so as to construct a layered representation of the image. Promising results that agree with human’s intuition are obtained, showing the potential of our approach in image retrieval and other related aspects of image management.
URI: http://hdl.handle.net/10397/26940
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

Page view(s)

40
Last Week
5
Last month
Checked on Aug 13, 2017

Google ScholarTM

Check



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