Multichannel image identification and restoration based on the EM algorithm and cross-validation

Aggelos K. Katsaggelos; Nikolas P. Galatsanos; Kuen Tsair Lay; Wenwu Zhu

Multichannel image identification and restoration based on the EM algorithm and cross-validation

Aggelos K. Katsaggelos^*, Nikolas P. Galatsanos, Kuen Tsair Lay, Wenwu Zhu

^*Corresponding author for this work

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Scopus citations

Abstract

In this paper we address some of the main shortcomings of multi-channel (MC) linear restoration filters. The problem of restoring a MC image and simultaneously estimating the MC power spectrum of the image and the noise, required by linear minimum mean squared error (LMMSE) filters is investigated, using the expectation-maximization (EM) algorithm. Second, the problem of estimating, the regularization parameters and operator, required by regularized least-squares (RLS) MC restoration filters is investigated using the cross-validation (CV) function. Furthermore, a novel representation of MC signal processing is introduced. This notation leads to a more natural extension of single-channel (SC) signal processing algorithms to the MC case and yields a new class of matrices which we call semi-block- circulant (SBC) matrices. The properties of these matrices are examined and a family of new efficient algorithms is developed for the computation of the MC EM and CV functions.

Original language	English (US)
Title of host publication	Proceedings of SPIE - The International Society for Optical Engineering
Publisher	Publ by Int Soc for Optical Engineering
Pages	147-158
Number of pages	12
ISBN (Print)	0819409405
State	Published - Dec 1 1992
Event	Inverse Problems in Scattering and Imaging - San Diego, CA, USA Duration: Jul 20 1992 → Jul 22 1992

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	1767
ISSN (Print)	0277-786X

Other

Other	Inverse Problems in Scattering and Imaging
City	San Diego, CA, USA
Period	7/20/92 → 7/22/92

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

Access to Document

Fingerprint Dive into the research topics of 'Multichannel image identification and restoration based on the EM algorithm and cross-validation'. Together they form a unique fingerprint.

Cite this

Katsaggelos, A. K., Galatsanos, N. P., Lay, K. T., & Zhu, W. (1992). Multichannel image identification and restoration based on the EM algorithm and cross-validation. In Proceedings of SPIE - The International Society for Optical Engineering (pp. 147-158). (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 1767). Publ by Int Soc for Optical Engineering.

Katsaggelos, Aggelos K. ; Galatsanos, Nikolas P. ; Lay, Kuen Tsair ; Zhu, Wenwu. / Multichannel image identification and restoration based on the EM algorithm and cross-validation. Proceedings of SPIE - The International Society for Optical Engineering. Publ by Int Soc for Optical Engineering, 1992. pp. 147-158 (Proceedings of SPIE - The International Society for Optical Engineering).

@inproceedings{25077de7d4df47f4b73a547ba9370029,

title = "Multichannel image identification and restoration based on the EM algorithm and cross-validation",

abstract = "In this paper we address some of the main shortcomings of multi-channel (MC) linear restoration filters. The problem of restoring a MC image and simultaneously estimating the MC power spectrum of the image and the noise, required by linear minimum mean squared error (LMMSE) filters is investigated, using the expectation-maximization (EM) algorithm. Second, the problem of estimating, the regularization parameters and operator, required by regularized least-squares (RLS) MC restoration filters is investigated using the cross-validation (CV) function. Furthermore, a novel representation of MC signal processing is introduced. This notation leads to a more natural extension of single-channel (SC) signal processing algorithms to the MC case and yields a new class of matrices which we call semi-block- circulant (SBC) matrices. The properties of these matrices are examined and a family of new efficient algorithms is developed for the computation of the MC EM and CV functions.",

author = "Katsaggelos, {Aggelos K.} and Galatsanos, {Nikolas P.} and Lay, {Kuen Tsair} and Wenwu Zhu",

year = "1992",

month = dec,

day = "1",

language = "English (US)",

isbn = "0819409405",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "Publ by Int Soc for Optical Engineering",

pages = "147--158",

booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",

note = "Inverse Problems in Scattering and Imaging ; Conference date: 20-07-1992 Through 22-07-1992",

}

Katsaggelos, AK, Galatsanos, NP, Lay, KT & Zhu, W 1992, Multichannel image identification and restoration based on the EM algorithm and cross-validation. in Proceedings of SPIE - The International Society for Optical Engineering. Proceedings of SPIE - The International Society for Optical Engineering, vol. 1767, Publ by Int Soc for Optical Engineering, pp. 147-158, Inverse Problems in Scattering and Imaging, San Diego, CA, USA, 7/20/92.

Multichannel image identification and restoration based on the EM algorithm and cross-validation. / Katsaggelos, Aggelos K.; Galatsanos, Nikolas P.; Lay, Kuen Tsair; Zhu, Wenwu.

Proceedings of SPIE - The International Society for Optical Engineering. Publ by Int Soc for Optical Engineering, 1992. p. 147-158 (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 1767).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Multichannel image identification and restoration based on the EM algorithm and cross-validation

AU - Katsaggelos, Aggelos K.

AU - Galatsanos, Nikolas P.

AU - Lay, Kuen Tsair

AU - Zhu, Wenwu

PY - 1992/12/1

Y1 - 1992/12/1

N2 - In this paper we address some of the main shortcomings of multi-channel (MC) linear restoration filters. The problem of restoring a MC image and simultaneously estimating the MC power spectrum of the image and the noise, required by linear minimum mean squared error (LMMSE) filters is investigated, using the expectation-maximization (EM) algorithm. Second, the problem of estimating, the regularization parameters and operator, required by regularized least-squares (RLS) MC restoration filters is investigated using the cross-validation (CV) function. Furthermore, a novel representation of MC signal processing is introduced. This notation leads to a more natural extension of single-channel (SC) signal processing algorithms to the MC case and yields a new class of matrices which we call semi-block- circulant (SBC) matrices. The properties of these matrices are examined and a family of new efficient algorithms is developed for the computation of the MC EM and CV functions.

AB - In this paper we address some of the main shortcomings of multi-channel (MC) linear restoration filters. The problem of restoring a MC image and simultaneously estimating the MC power spectrum of the image and the noise, required by linear minimum mean squared error (LMMSE) filters is investigated, using the expectation-maximization (EM) algorithm. Second, the problem of estimating, the regularization parameters and operator, required by regularized least-squares (RLS) MC restoration filters is investigated using the cross-validation (CV) function. Furthermore, a novel representation of MC signal processing is introduced. This notation leads to a more natural extension of single-channel (SC) signal processing algorithms to the MC case and yields a new class of matrices which we call semi-block- circulant (SBC) matrices. The properties of these matrices are examined and a family of new efficient algorithms is developed for the computation of the MC EM and CV functions.

UR - http://www.scopus.com/inward/record.url?scp=0026990134&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0026990134&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0026990134

SN - 0819409405

T3 - Proceedings of SPIE - The International Society for Optical Engineering

SP - 147

EP - 158

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - Publ by Int Soc for Optical Engineering

T2 - Inverse Problems in Scattering and Imaging

Y2 - 20 July 1992 through 22 July 1992

ER -