Neural Approaches to Image Compression/Decompression Using PCA based Learning Algorithms

Luminita State, Catalina Cocianu, Panayiotis Vlamos, Doru Constantin

2008

Abstract

Principal Component Analysis is a well-known statistical method for feature extraction, data compression and multivariate data projection. Aiming to obtain a guideline for choosing a proper method for a specific application we developed a series of simulations on some the most currently used PCA algorithms as GHA, Sanger variant of GHA and APEX. The paper reports the conclusions experimentally derived on the convergence rates and their corresponding efficiency for specific image processing tasks.

References

  1. Chatterjee, C., Roychowdhury, V.P., Chong, E.K.P.: On Relative Convergence Properties of PCA Algorithms, IEEE Trans. on Neural Networks, vol.9,no.2 (1998)
  2. Diamantaras, K.I., Kung, S.Y.: Principal Component Neural Networks: theory and applications, John Wiley &Sons, (1996)
  3. Haykin, S., Neural Networks A Comprehensive Foundation, Prentice Hall,Inc. (1999)
  4. Hastie, T., Tibshirani, R., Friedman,J.: The Elements of Statistical Learning Data Mining, Inference, and Prediction, Springer (2001)
  5. Kushner, H.J., Clark, D.S.: Stochastic Approximation Methods for Constrained and Unconstrained Systems, Springer Verlag (1978)
  6. J. Karhunen, E. Oja: New Methods for Stochastic Approximations of Truncated KarhunenLoeve Expansions, Proc. 6th Intl. Conf. on Pattern Recognition, Springer Verlag (1982)
  7. Sanger, T.D.: An Optimality Principle for Unsupervised Learning, Advances in Neural Information Systems, ed. D.S. Touretzky, Morgan Kaufmann (1989)
Download


Paper Citation


in Harvard Style

State L., Cocianu C., Vlamos P. and Constantin D. (2008). Neural Approaches to Image Compression/Decompression Using PCA based Learning Algorithms . In Proceedings of the 8th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2008) ISBN 978-989-8111-42-5, pages 187-192. DOI: 10.5220/0001728701870192


in Bibtex Style

@conference{pris08,
author={Luminita State and Catalina Cocianu and Panayiotis Vlamos and Doru Constantin},
title={Neural Approaches to Image Compression/Decompression Using PCA based Learning Algorithms},
booktitle={Proceedings of the 8th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2008)},
year={2008},
pages={187-192},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001728701870192},
isbn={978-989-8111-42-5},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 8th International Workshop on Pattern Recognition in Information Systems - Volume 1: PRIS, (ICEIS 2008)
TI - Neural Approaches to Image Compression/Decompression Using PCA based Learning Algorithms
SN - 978-989-8111-42-5
AU - State L.
AU - Cocianu C.
AU - Vlamos P.
AU - Constantin D.
PY - 2008
SP - 187
EP - 192
DO - 10.5220/0001728701870192