Please use this identifier to cite or link to this item: https://doi.org/10.1023/A:1009692213403
Title: Approximating content-based object-level image retrieval
Authors: Hsu, W. 
Chua, T.S. 
Pung, H.K. 
Issue Date: 2000
Source: Hsu, W., Chua, T.S., Pung, H.K. (2000). Approximating content-based object-level image retrieval. Multimedia Tools and Applications 12 (1) : 59-79. ScholarBank@NUS Repository. https://doi.org/10.1023/A:1009692213403
Abstract: Object-level image retrieval is an active area of research. Given an image, a human observer does not see random dots of colors. Rather, he/she observes familiar objects in the image. Therefore, to make image retrieval more user-friendly and more effective and efficient, object-level image retrieval technique is necessary. Unfortunately, images today are mostly represented as 2D arrays of pixels values. The object-level semantics of the images are not captured. Researchers try to overcome this problem by attempting to deduce the object-level semantics through additional information such as the motion vectors in the case of video clips. Some success stories have been reported. However, deducing object-level semantics from still images is still a difficult problem. In this paper, we propose a 'color-spatial' approach to approximate object-level image retrieval. The color and spatial information of the principle components of an object are estimated. The technique involves three steps: the selection of the principle component colors, the analysis of spatial information of the selected colors, and the retrieval process based on the color-spatial information. Two color histograms are used to aid in the process of color selection. After deriving the set of representative colors, spatial knowledge of the selected colors is obtained using a maximum entropy discretization with event covering method. A retrieval process is formulated to make use of the spatial knowledge for retrieving relevant images. A prototype image retrieved tool has been implemented on the Unix system. It is tested on two image database consisting of 260 images and 11,111 images respectively. The results show that the 'color-spatial' approach is able to retrieve similar objects with much better precision than the sole color-based retrieval methods.
Source Title: Multimedia Tools and Applications
URI: http://scholarbank.nus.edu.sg/handle/10635/39065
ISSN: 13807501
DOI: 10.1023/A:1009692213403
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

14
checked on Jan 16, 2018

WEB OF SCIENCETM
Citations

9
checked on Dec 11, 2017

Page view(s)

58
checked on Jan 20, 2018

Google ScholarTM

Check

Altmetric


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