Improving similarity measures of histograms using smoothing projections

Authors:
Joni-Kristian Kamarainen;Ville Kyrki;Jarmo Ilonen;Heikki Kälviäinen
Affiliations:
Laboratory of Information Processing, Department of Information Technology, Lappeenranta University of Technology, P.O. Box 20, FIN-53851 Lappeenranta, Finland;Laboratory of Information Processing, Department of Information Technology, Lappeenranta University of Technology, P.O. Box 20, FIN-53851 Lappeenranta, Finland;Laboratory of Information Processing, Department of Information Technology, Lappeenranta University of Technology, P.O. Box 20, FIN-53851 Lappeenranta, Finland;Laboratory of Information Processing, Department of Information Technology, Lappeenranta University of Technology, P.O. Box 20, FIN-53851 Lappeenranta, Finland
Venue:
Pattern Recognition Letters
Year:
2003

Citing 9
Cited 12

Toward memory-based reasoning

Communications of the ACM - Special issue on parallelism
Learning Based on Conceptual Distance

IEEE Transactions on Pattern Analysis and Machine Intelligence
Toward Improved Ranking Metrics

IEEE Transactions on Pattern Analysis and Machine Intelligence
Feature normalization and likelihood-based similarity measures for image retrieval

Pattern Recognition Letters - Special issue on image/video indexing and retrieval
Varying similarity metrics in visual information retrieval

Pattern Recognition Letters - Special issue on image/video indexing and retrieval
Unsupervised Feature Selection Using Feature Similarity

IEEE Transactions on Pattern Analysis and Machine Intelligence
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Efficient Color Histogram Indexing for Quadratic Form Distance Functions

IEEE Transactions on Pattern Analysis and Machine Intelligence
Improved heterogeneous distance functions

Journal of Artificial Intelligence Research

Feature representation and discrimination based on Gaussian mixture model probability densities-Practices and algorithms

Pattern Recognition
Signatures versus histograms: Definitions, distances and algorithms

Pattern Recognition
A modified Gabor function for content based image retrieval

Pattern Recognition Letters
Combining neighbourhood-based and histogram similarity measures for the design of image quality measures

Image and Vision Computing
A new similarity measure for histogram comparison and its application in time series analysis

Pattern Recognition Letters
Combining similarity measures in content-based image retrieval

Pattern Recognition Letters
Feature selection based-on genetic algorithm for image annotation

Knowledge-Based Systems
LCS-Hist: taming massive high-dimensional data cube compression

Proceedings of the 12th International Conference on Extending Database Technology: Advances in Database Technology
Bayesian model checking for multivariate outcome data

Computational Statistics & Data Analysis
Dynamic Adaboost learning with feature selection based on parallel genetic algorithm for image annotation

Knowledge-Based Systems
A fast and exact modulo-distance between histograms

SSPR'06/SPR'06 Proceedings of the 2006 joint IAPR international conference on Structural, Syntactic, and Statistical Pattern Recognition
A fast distance between histograms

CIARP'05 Proceedings of the 10th Iberoamerican Congress conference on Progress in Pattern Recognition, Image Analysis and Applications

Quantified Score

Hi-index	0.10

Visualization

Abstract

Selection of a proper similarity measure is an essential consideration for a success of many methods. In this study, similarity measures are analyzed in the context of ordered histogram type data, such as gray-level histograms of digital images or color spectra. Furthermore, the performance of the studied similarity measures can be improved using a smoothing projection, called neighbor-bank projection. Especially, with distance functions utilizing statistical properties of data, e.g., the Mahalanobis distance, a significant improvement was achieved in the classification experiments on real data sets, resulting from the use of a priori information related to ordered data. The proposed projection seems also to be applicable for dimensional reduction of histograms and to represent sparse data in a more tight form in the projection subspace.