Multiple Description Pattern Analysis: Robustness to Misclassification Using Local Discriminant Frame Expansions

Widhyakorn ASDORNWISED; Somchai JITAPUNKUL

doi:10.1093/ietisy/e88-d.10.2296

Multiple Description Pattern Analysis: Robustness to Misclassification Using Local Discriminant Frame Expansions

Widhyakorn ASDORNWISED, Somchai JITAPUNKUL

Full Text Views

0

Cite this

Summary :

In this paper, a source coding model for learning multiple concept descriptions of data is proposed. Our source coding model is based on the concept of transmitting data over multiple channels, called multiple description (MD) coding. In particular, frame expansions have been used in our MD coding models for pattern classification. Using this model, there are several interesting properties within a class of multiple classifier algorithms that share with our proposed scheme. Generalization of the MD view under an extension of local discriminant basis towards the theory of frames allows the formulation of a generalized class of low-complexity learning algorithms applicable to high-dimensional pattern classification. To evaluate this approach, performance results for automatic target recognition (ATR) are presented for synthetic aperture radar (SAR) images from the MSTAR public release data set. From the experimental results, our approach outperforms state-of-the-art methods such as conditional Gaussian signal model, Adaboost, and ECOC-SVM.

Publication: IEICE TRANSACTIONS on Information Vol.E88-D No.10 pp.2296-2307

Publication Date: 2005/10/01

Publicized

Online ISSN

DOI: 10.1093/ietisy/e88-d.10.2296

Type of Manuscript: Special Section PAPER (Special Section on Image Recognition and Understanding)

Category

Cite this

Copy

Widhyakorn ASDORNWISED, Somchai JITAPUNKUL, "Multiple Description Pattern Analysis: Robustness to Misclassification Using Local Discriminant Frame Expansions" in IEICE TRANSACTIONS on Information, vol. E88-D, no. 10, pp. 2296-2307, October 2005, doi: 10.1093/ietisy/e88-d.10.2296.
Abstract: In this paper, a source coding model for learning multiple concept descriptions of data is proposed. Our source coding model is based on the concept of transmitting data over multiple channels, called multiple description (MD) coding. In particular, frame expansions have been used in our MD coding models for pattern classification. Using this model, there are several interesting properties within a class of multiple classifier algorithms that share with our proposed scheme. Generalization of the MD view under an extension of local discriminant basis towards the theory of frames allows the formulation of a generalized class of low-complexity learning algorithms applicable to high-dimensional pattern classification. To evaluate this approach, performance results for automatic target recognition (ATR) are presented for synthetic aperture radar (SAR) images from the MSTAR public release data set. From the experimental results, our approach outperforms state-of-the-art methods such as conditional Gaussian signal model, Adaboost, and ECOC-SVM.
URL: https://global.ieice.org/en_transactions/information/10.1093/ietisy/e88-d.10.2296/_p

Copy

@ARTICLE{e88-d_10_2296,
author={Widhyakorn ASDORNWISED, Somchai JITAPUNKUL, },
journal={IEICE TRANSACTIONS on Information},
title={Multiple Description Pattern Analysis: Robustness to Misclassification Using Local Discriminant Frame Expansions},
year={2005},
volume={E88-D},
number={10},
pages={2296-2307},
abstract={In this paper, a source coding model for learning multiple concept descriptions of data is proposed. Our source coding model is based on the concept of transmitting data over multiple channels, called multiple description (MD) coding. In particular, frame expansions have been used in our MD coding models for pattern classification. Using this model, there are several interesting properties within a class of multiple classifier algorithms that share with our proposed scheme. Generalization of the MD view under an extension of local discriminant basis towards the theory of frames allows the formulation of a generalized class of low-complexity learning algorithms applicable to high-dimensional pattern classification. To evaluate this approach, performance results for automatic target recognition (ATR) are presented for synthetic aperture radar (SAR) images from the MSTAR public release data set. From the experimental results, our approach outperforms state-of-the-art methods such as conditional Gaussian signal model, Adaboost, and ECOC-SVM.},
keywords={},
doi={10.1093/ietisy/e88-d.10.2296},
ISSN={},
month={October},}

Copy

TY - JOUR
TI - Multiple Description Pattern Analysis: Robustness to Misclassification Using Local Discriminant Frame Expansions
T2 - IEICE TRANSACTIONS on Information
SP - 2296
EP - 2307
AU - Widhyakorn ASDORNWISED
AU - Somchai JITAPUNKUL
PY - 2005
DO - 10.1093/ietisy/e88-d.10.2296
JO - IEICE TRANSACTIONS on Information
SN -
VL - E88-D
IS - 10
JA - IEICE TRANSACTIONS on Information
Y1 - October 2005
AB - In this paper, a source coding model for learning multiple concept descriptions of data is proposed. Our source coding model is based on the concept of transmitting data over multiple channels, called multiple description (MD) coding. In particular, frame expansions have been used in our MD coding models for pattern classification. Using this model, there are several interesting properties within a class of multiple classifier algorithms that share with our proposed scheme. Generalization of the MD view under an extension of local discriminant basis towards the theory of frames allows the formulation of a generalized class of low-complexity learning algorithms applicable to high-dimensional pattern classification. To evaluate this approach, performance results for automatic target recognition (ATR) are presented for synthetic aperture radar (SAR) images from the MSTAR public release data set. From the experimental results, our approach outperforms state-of-the-art methods such as conditional Gaussian signal model, Adaboost, and ECOC-SVM.
ER -