Image restoration using fast modified reduced update Kalman filter

Wen-Rong Wu; Amlan Kundu

doi:10.1109/ICSYSE.1990.203215

Image restoration using fast modified reduced update Kalman filter

Wen-Rong Wu^*, Amlan Kundu

^*Corresponding author for this work

Department of Electrical and Computer Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract

Some modifications to the reduced update Kalman filter (RUKF) as applied to the filtering of images corrupted by additive noise are proposed. The computational complexity of RUKF is reduced by reducing the state dimensionality. The RUKF is modified using the score-function-based approach to accommodate the non-Gaussian noise. The image is modeled as a nonstationary mean and stationary variance autoregressive Gaussian process. It is shown that the stationary variance assumption is reasonable if the nonstationary mean is computed by means of an edge and detail preserving spatial filter. Such a filter is described.

Original language	English
Pages	547-550
Number of pages	4
DOIs	https://doi.org/10.1109/ICSYSE.1990.203215
State	Published - 9 Aug 1990
Event	Proceedings of the 1990 IEEE International Conference on Systems Engineering - Pittsburgh, PA, USA Duration: 9 Aug 1990 → 11 Aug 1990

Conference

Conference	Proceedings of the 1990 IEEE International Conference on Systems Engineering
City	Pittsburgh, PA, USA
Period	9/08/90 → 11/08/90

Access to Document

10.1109/ICSYSE.1990.203215

Cite this

@conference{0e39bcd5254b4250b9158bf7daf5f66f,

title = "Image restoration using fast modified reduced update Kalman filter",

abstract = "Some modifications to the reduced update Kalman filter (RUKF) as applied to the filtering of images corrupted by additive noise are proposed. The computational complexity of RUKF is reduced by reducing the state dimensionality. The RUKF is modified using the score-function-based approach to accommodate the non-Gaussian noise. The image is modeled as a nonstationary mean and stationary variance autoregressive Gaussian process. It is shown that the stationary variance assumption is reasonable if the nonstationary mean is computed by means of an edge and detail preserving spatial filter. Such a filter is described.",

author = "Wen-Rong Wu and Amlan Kundu",

year = "1990",

month = aug,

day = "9",

doi = "10.1109/ICSYSE.1990.203215",

language = "English",

pages = "547--550",

note = "Proceedings of the 1990 IEEE International Conference on Systems Engineering ; Conference date: 09-08-1990 Through 11-08-1990",

}

TY - CONF

T1 - Image restoration using fast modified reduced update Kalman filter

AU - Wu, Wen-Rong

AU - Kundu, Amlan

PY - 1990/8/9

Y1 - 1990/8/9

N2 - Some modifications to the reduced update Kalman filter (RUKF) as applied to the filtering of images corrupted by additive noise are proposed. The computational complexity of RUKF is reduced by reducing the state dimensionality. The RUKF is modified using the score-function-based approach to accommodate the non-Gaussian noise. The image is modeled as a nonstationary mean and stationary variance autoregressive Gaussian process. It is shown that the stationary variance assumption is reasonable if the nonstationary mean is computed by means of an edge and detail preserving spatial filter. Such a filter is described.

AB - Some modifications to the reduced update Kalman filter (RUKF) as applied to the filtering of images corrupted by additive noise are proposed. The computational complexity of RUKF is reduced by reducing the state dimensionality. The RUKF is modified using the score-function-based approach to accommodate the non-Gaussian noise. The image is modeled as a nonstationary mean and stationary variance autoregressive Gaussian process. It is shown that the stationary variance assumption is reasonable if the nonstationary mean is computed by means of an edge and detail preserving spatial filter. Such a filter is described.

UR - http://www.scopus.com/inward/record.url?scp=0025536533&partnerID=8YFLogxK

U2 - 10.1109/ICSYSE.1990.203215

DO - 10.1109/ICSYSE.1990.203215

M3 - Paper

AN - SCOPUS:0025536533

SP - 547

EP - 550

T2 - Proceedings of the 1990 IEEE International Conference on Systems Engineering

Y2 - 9 August 1990 through 11 August 1990

ER -

Image restoration using fast modified reduced update Kalman filter

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this