Constructing colour image databases using BTC for efficient storage and effective retrieval

Author

Qiu, G.

Author_Institution

Sch. of Comput. Sci., Nottingham Univ., UK

fYear

2001

fDate

2001

Firstpage

364

Lastpage

367

Abstract

We introduce a simple image coding method, the block truncation coding (BTC) technique, as a novel approach to the construction of colour image databases. It is shown that BTC can not only be used to compress images, thus achieving storage efficiency, but the BTC codes can also be used directly to construct image features for effective image retrieval. From the BTC code we have developed an image feature termed the BTC colour co-occurrence matrix (BCCM) as an effective measure of image contents. Experimental results are presented to show that BCCM is comparable to state of the art techniques, such as color correlogram, in image retrieval

Keywords

block codes; colour graphics; data compression; image coding; image retrieval; visual databases; BCCM; BTC codes; BTC colour co-occurrence matrix; block truncation coding technique; color correlogram; colour image database construction; image coding method; image compression; image contents; image feature; image features; image retrieval; storage efficiency; Color; Image coding; Image databases; Image processing; Image retrieval; Image storage; Indexing; Information retrieval; Spatial databases; Visual databases;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Multimedia, Video and Speech Processing, 2001. Proceedings of 2001 International Symposium on

Conference_Location

Hong Kong

Print_ISBN

962-85766-2-3

Type

conf

DOI

10.1109/ISIMP.2001.925409

Filename

925409