A set of novel textural features based on 3D co-occurrence matrix for content-based image retrieval

Author

Dacheng, Tao ; Xuelong, Li ; Yuan, YUAN ; Nenghai, Yu ; Zhengkai, Liu ; Xiao-ou, TANG

Author_Institution

Dept. of Electron. Eng. & Inf. Sci., Univ. of Sci. & Technol. of China, Hefei, China

Volume

2

fYear

2002

fDate

8-11 July 2002

Firstpage

1403

Abstract

This paper presents a set of novel texture features for content-based image retrieval (CBIR). CBIR require new algorithms for the automated extraction and indexing of salient image features, while texture features provide one important cue for the visual perception and discrimination of image content, so they are always one of the most popular representations in CBIR. The proposed features are based on the 3D co-occurrence matrix, and preliminary results show they are much better than traditional texture features. In this paper, experiments are made on an image database with 25000 images.

Keywords

content-based retrieval; database indexing; image retrieval; image texture; visual databases; 3D co-occurrence matrix; automated salient image feature extraction; automated salient image feature indexing; content-based image retrieval; texture features; visual discrimination; visual perception; Content based retrieval; Data mining; Feature extraction; Image databases; Image retrieval; Indexing; Information retrieval; Shape; Spatial databases; Visual perception;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Fusion, 2002. Proceedings of the Fifth International Conference on

Conference_Location

Annapolis, MD, USA

Print_ISBN

0-9721844-1-4

Type

conf

DOI

10.1109/ICIF.2002.1020979

Filename

1020979