Content-Based Image Retrieval System using Keyword Mapping and Color Features


The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 5, No. 10, pp. 2498-2511, Oct. 1998
10.3745/KIPSTE.1998.5.10.2498,   PDF Download:

Abstract

In this paper, we propose and implement a new method of content-based image retrieval using the color features which are mapped from color position-description keywords and color keywords. Color keywords are defined by color segment primitives using color spatial information. Position-description keywords are defined by position segment primitives using color regional information. Each color keyword primitive is mapped into the color features of images, and after it is compared with the color features of 6x6 blocks of reference images in feature database, candidates are retrieved in accordance with the similarity order. The retrieval accuracy was evaluated through retrieval experiments about 3,800 images. The experimental results of the image retrieval are presented to illustrate the usefulness of this approach that demonstrates the average recall/precision, 0.72/0.80 in content-based image data retrieval.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
C. K. Ho and C. H. Sub, "Content-Based Image Retrieval System using Keyword Mapping and Color Features," The Transactions of the Korea Information Processing Society (1994 ~ 2000), vol. 5, no. 10, pp. 2498-2511, 1998. DOI: 10.3745/KIPSTE.1998.5.10.2498.

[ACM Style]
Choi Ki Ho and Choi Hyun Sub. 1998. Content-Based Image Retrieval System using Keyword Mapping and Color Features. The Transactions of the Korea Information Processing Society (1994 ~ 2000), 5, 10, (1998), 2498-2511. DOI: 10.3745/KIPSTE.1998.5.10.2498.