Efficient and fast multi layer statistical approach for colour based image retrieval


Odeh, , JQ and Johari,, R and Othman,, M and Ahmad,, F (2003) Efficient and fast multi layer statistical approach for colour based image retrieval. DIGITAL LIBRARIES: TECHNOLOGY AND MANAGEMENT OF INDIGENOUS KNOWLEDGE FOR GLOBAL ACCESS, 2911 . pp. 134-148. ISSN 0302-9743

Full text not available from this repository.


In this paper a new efficient and fast technique for colour-based image retrieval is presented. The technique is based on utilizing singular feature in a multi layer system (SFMLSA). The colour features are extracted from image query and images database then distance measure based on city block is used to filter a set of images in each layer. Our approach attempts to overcome the computational complexity of applying bin-to-bin comparison as a multi dimensional feature vectors in the colour histogram approach. Furthermore, the proposed technique eliminate the needs of using the weight matrix, which is usually applied when more than one feature is combined together to judge on the similarity. This needs pre-knowledge of the conditions under which the images are captured. Throughout this paper a comparative study is carried out to examine the performance of the proposed approach with reference to an information theoretic approach using entropy as a discriminator for huge image database. Moreover, we examined the possibility of using the eigenvalues as a discernment feature for colour images, so we developed the necessary algorithms to test this approach. Different database sets has been used and the related algorithms are presented.

Item Type: Article
Subjects: Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 24 Aug 2011 00:12
Last Modified: 24 Aug 2011 00:12
URII: http://shdl.mmu.edu.my/id/eprint/2596


Downloads per month over past year

View ItemEdit (login required)