Image classification via LZ78 based string kernel: a comparative study

Authors:
Ming Li;Yanong Zhu
Affiliations:
School of Computing Sciences, University of East Anglia, Norwich, UK;School of Computing Sciences, University of East Anglia, Norwich, UK
Venue:
PAKDD'06 Proceedings of the 10th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining
Year:
2006

Citing 6
Cited 5

An introduction to Kolmogorov complexity and its applications (2nd ed.)

An introduction to Kolmogorov complexity and its applications (2nd ed.)
The similarity metric

SODA '03 Proceedings of the fourteenth annual ACM-SIAM symposium on Discrete algorithms
Towards parameter-free data mining

Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
A Robust Approach to Sequence Classification

ICTAI '05 Proceedings of the 17th IEEE International Conference on Tools with Artificial Intelligence
An LZ78 based string kernel

ADMA'05 Proceedings of the First international conference on Advanced Data Mining and Applications
Clustering by compression

IEEE Transactions on Information Theory

Dictionary based color image retrieval

Journal of Visual Communication and Image Representation
Modelling Image Complexity by Independent Component Analysis, with Application to Content-Based Image Retrieval

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part II
Detecting visually similar Web pages: Application to phishing detection

ACM Transactions on Internet Technology (TOIT)
Efficient LZ78 factorization of grammar compressed text

SPIRE'12 Proceedings of the 19th international conference on String Processing and Information Retrieval
Dictionary-based color image retrieval using multiset theory

Journal of Visual Communication and Image Representation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Normalized Information Distance (NID) [1] is a general-purpose similarity metric based on the concept of Kolmogorov Complexity. We have developed this notion into a valid kernel distance, called LZ78-based string kernel [2] and have shown that it can be used effectively for a variety of 1D sequence classification tasks [3]. In this paper, we further demonstrate its applicability on 2D images. We report experiments with our technique on two real datasets: (i) a collection of real-life photographs and (ii) a collection of medical diagnostic images from Magnetic Resonance (MR) data. The classification results are compared with those of the original similarity metric (i.e. NID) and several conventional classification algorithms. In all cases, the proposed kernel approach demonstrates better or equivalent performance when compared with other candidate methods but with lower computational overhead.