A novel BP-based image retrieval system

Author

Han, Jun-Hua ; Huang, De-Shuang

Author_Institution

Inst. of Intelligent Machines, Chinese Acad. of Sci., Anhui, China

fYear

2005

fDate

23-26 May 2005

Firstpage

1557

Abstract

This paper presents a novel BP-based image retrieval (BPBIR) system, which is based on the observation that the images users need are often similar to a set of images with the same conception instead of one query image and the assumption that there is a nonlinear relationship between different features. If user is not satisfied with the retrieved result, a relevance feedback method is used to enhance the performance of the proposed system by changing the weights of the BP neural networks. In addition, we describe a 5-block division method to give rough information on the spatial color composition. Finally, we compare the performance of the proposed system with other systems. Experimental results show the efficacy of the proposed system.

Keywords

backpropagation; image colour analysis; image retrieval; neural nets; relevance feedback; 5-block division method; BP neural network weights; BP-based image retrieval; BPBIR; performance; relevance feedback; rough information; spatial color composition; Content based retrieval; Hardware; Image databases; Image generation; Image retrieval; Information retrieval; Machine intelligence; Military computing; Neural networks; Neurofeedback;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2005. ISCAS 2005. IEEE International Symposium on

Print_ISBN

0-7803-8834-8

Type

conf

DOI

10.1109/ISCAS.2005.1464898

Filename

1464898