Multiresolution wavelet analysis based feature extraction for neural network classification

C. H. Chen, Gwo-Giun Lee

Research output: Contribution to conferencePaper

7 Citations (Scopus)

Abstract

In this paper we introduce a novel feature extraction scheme as a preprocessor for artificial neural network (ANN) classification. We have shown that the feature extraction scheme implemented via a non-stationary Gaussian Markov random field (GMRF) based on a multiresolution wavelet framework can provide effective features for both the ANN and Fuzzy C-Means (FCM) classification. In our experiment with natural textures and real world digital mammography, each region of the tested images is assumed to be a different class. A label field with each region or class being represented by the same grayscale was then found by the back propagation neural network (BPNN) and FCM clustering algorithm using the extracted discriminatory features. Further enhancement of the segmented result was achieved via Bayesian learning. The formulation of this maximum a posteriori (MAP) estimator was based on the Gibbs prior assumption which is especially appropriate for modeling real world mammograms. Although being estimated by constrained optimization, the MAP estimator can also be found from neural networks such as the Boltzman and the Mean-field-theory machines.

Original languageEnglish
Pages1416-1421
Number of pages6
Publication statusPublished - 1996 Jan 1
EventProceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) - Washington, DC, USA
Duration: 1996 Jun 31996 Jun 6

Other

OtherProceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4)
CityWashington, DC, USA
Period96-06-0396-06-06

Fingerprint

Wavelet analysis
Feature extraction
Neural networks
Mean field theory
Mammography
Constrained optimization
Backpropagation
Clustering algorithms
Labels
Textures
Experiments

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Chen, C. H., & Lee, G-G. (1996). Multiresolution wavelet analysis based feature extraction for neural network classification. 1416-1421. Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, .
Chen, C. H. ; Lee, Gwo-Giun. / Multiresolution wavelet analysis based feature extraction for neural network classification. Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, .6 p.
@conference{0e35bb4f7eb84b948d859237f562d08b,
title = "Multiresolution wavelet analysis based feature extraction for neural network classification",
abstract = "In this paper we introduce a novel feature extraction scheme as a preprocessor for artificial neural network (ANN) classification. We have shown that the feature extraction scheme implemented via a non-stationary Gaussian Markov random field (GMRF) based on a multiresolution wavelet framework can provide effective features for both the ANN and Fuzzy C-Means (FCM) classification. In our experiment with natural textures and real world digital mammography, each region of the tested images is assumed to be a different class. A label field with each region or class being represented by the same grayscale was then found by the back propagation neural network (BPNN) and FCM clustering algorithm using the extracted discriminatory features. Further enhancement of the segmented result was achieved via Bayesian learning. The formulation of this maximum a posteriori (MAP) estimator was based on the Gibbs prior assumption which is especially appropriate for modeling real world mammograms. Although being estimated by constrained optimization, the MAP estimator can also be found from neural networks such as the Boltzman and the Mean-field-theory machines.",
author = "Chen, {C. H.} and Gwo-Giun Lee",
year = "1996",
month = "1",
day = "1",
language = "English",
pages = "1416--1421",
note = "Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4) ; Conference date: 03-06-1996 Through 06-06-1996",

}

Chen, CH & Lee, G-G 1996, 'Multiresolution wavelet analysis based feature extraction for neural network classification' Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, 96-06-03 - 96-06-06, pp. 1416-1421.

Multiresolution wavelet analysis based feature extraction for neural network classification. / Chen, C. H.; Lee, Gwo-Giun.

1996. 1416-1421 Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Multiresolution wavelet analysis based feature extraction for neural network classification

AU - Chen, C. H.

AU - Lee, Gwo-Giun

PY - 1996/1/1

Y1 - 1996/1/1

N2 - In this paper we introduce a novel feature extraction scheme as a preprocessor for artificial neural network (ANN) classification. We have shown that the feature extraction scheme implemented via a non-stationary Gaussian Markov random field (GMRF) based on a multiresolution wavelet framework can provide effective features for both the ANN and Fuzzy C-Means (FCM) classification. In our experiment with natural textures and real world digital mammography, each region of the tested images is assumed to be a different class. A label field with each region or class being represented by the same grayscale was then found by the back propagation neural network (BPNN) and FCM clustering algorithm using the extracted discriminatory features. Further enhancement of the segmented result was achieved via Bayesian learning. The formulation of this maximum a posteriori (MAP) estimator was based on the Gibbs prior assumption which is especially appropriate for modeling real world mammograms. Although being estimated by constrained optimization, the MAP estimator can also be found from neural networks such as the Boltzman and the Mean-field-theory machines.

AB - In this paper we introduce a novel feature extraction scheme as a preprocessor for artificial neural network (ANN) classification. We have shown that the feature extraction scheme implemented via a non-stationary Gaussian Markov random field (GMRF) based on a multiresolution wavelet framework can provide effective features for both the ANN and Fuzzy C-Means (FCM) classification. In our experiment with natural textures and real world digital mammography, each region of the tested images is assumed to be a different class. A label field with each region or class being represented by the same grayscale was then found by the back propagation neural network (BPNN) and FCM clustering algorithm using the extracted discriminatory features. Further enhancement of the segmented result was achieved via Bayesian learning. The formulation of this maximum a posteriori (MAP) estimator was based on the Gibbs prior assumption which is especially appropriate for modeling real world mammograms. Although being estimated by constrained optimization, the MAP estimator can also be found from neural networks such as the Boltzman and the Mean-field-theory machines.

UR - http://www.scopus.com/inward/record.url?scp=0029727314&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029727314&partnerID=8YFLogxK

M3 - Paper

SP - 1416

EP - 1421

ER -

Chen CH, Lee G-G. Multiresolution wavelet analysis based feature extraction for neural network classification. 1996. Paper presented at Proceedings of the 1996 IEEE International Conference on Neural Networks, ICNN. Part 1 (of 4), Washington, DC, USA, .