Binary co-occurrence matrix in image database indexing

Authors:
Iivari Kunttu;Leena Lepistö;Juhani Rauhamaa;Ari Visa
Affiliations:
Tampere University of Technology, Institute of Signal Processing, Tampere, Finland;Tampere University of Technology, Institute of Signal Processing, Tampere, Finland;ABB Oy, Paper, Printing, Metals & Minerals, Automation, Helsinki, Finland;Tampere University of Technology, Institute of Signal Processing, Tampere, Finland
Venue:
SCIA'03 Proceedings of the 13th Scandinavian conference on Image analysis
Year:
2003

Citing 4
Cited 1

Principles of data mining

Principles of data mining
Modern Information Retrieval

Modern Information Retrieval
Image Indexing Using Color Correlograms

CVPR '97 Proceedings of the 1997 Conference on Computer Vision and Pattern Recognition (CVPR '97)
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)

Texture Description Through Histograms of Equivalent Patterns

Journal of Mathematical Imaging and Vision

Quantified Score

Hi-index	0.00

Visualization

Abstract

The use of the second order statistical measures has became popular in the image database indexing and retrieval. Unlike the common approach, image histogram, second order statistics like image correlogram and autocorrelogram consider also the spatial organization of the image colors or gray levels. Recently, correlograms and autocorrelograms have been widely used in the image database indexing. In this paper we present binary co-occurrence matrix, a new statistical measure for image indexing. This measure represents the footprint distribution of the co-occurrence matrix. Compared to image correlogram, this approach provides better retrieval accuracy at lower computational cost. We make retrieval experiments using two industrial image databases. These databases contain images collected from paper and metal manufacturing processes. In the experiments, we compare the retrieval performance of our approach to that of correlograms and autocorrelograms.