A color indexing scheme using two-Level clustering processing for effective and efficient image retrieval
Paper
Paper/Presentation Title | A color indexing scheme using two-Level clustering processing for effective and efficient image retrieval |
---|---|
Presentation Type | Paper |
Authors | Zhang, Ji (Author), Wang, Wei (Author) and Zhang, Sheng (Author) |
Editors | Arabnia, Hamid R. and Scime, A. |
Journal or Proceedings Title | Proceedings of the 2005 International Conference on Data Mining (DMIN'05) |
Number of Pages | 6 |
Year | 2005 |
Place of Publication | Las Vegas, Nevada, USA |
Conference/Event | 2005 International Conference on Data Mining (DMIN'05) |
Event Details | 2005 International Conference on Data Mining (DMIN'05) Parent International Conference on Data Mining Event Date Jun 2005 Event Location Las Vegas, United States of America |
Abstract | [Abstract]: In this paper, we present a clustering-based color indexing scheme for effective amd efficient image retrieval, which is essentially an exploration on the application of clustering technique to image retrieval. In our approach, the color features are clustered automatically using a color clustering algorithm twice (called two-level clustering processing) and two color feature summarizations are obtained, e.g. Local Color Centriods (LCCs) and Global Color Centroids (GCCs). Based upon LCCs and GCCs, a three-level R-tree is bulit for indexing database images and performing effecitve and efficient image retrieval. The experiments show that this indexing scheme is effective and efficient in performing image retrieval. |
Keywords | image indexing and retrieval, clustering, data mining, R-tree |
ANZSRC Field of Research 2020 | 469999. Other information and computing sciences not elsewhere classified |
Public Notes | No evidence of copyright restrictions. |
Byline Affiliations | University of Toronto, Canada |
Nanjing Normal University, China |
https://research.usq.edu.au/item/9z289/a-color-indexing-scheme-using-two-level-clustering-processing-for-effective-and-efficient-image-retrieval
Download files
1877
total views195
total downloads0
views this month0
downloads this month