Deconvolution of Irregularly Subsampled Images

Ahmed Karam-Eldaly, Yoann Altmann, Antonios Perperidis, Stephen McLaughlin

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Imaging technologies, such as coherent fibred bundle optical microscopy (FBOM), operate with irregularly-spaced sparse subsamples from their field of view. In this paper, we address the problem of data deconvolution for applications where the observed irregularly distributed samples are considered as a result of a convolution operator acting on original samples and corrupted by additive observation noise. We propose a hierarchical Bayesian model in which suitable prior distributions are assigned to the unknown model parameters, and compare two estimation strategies including Markov chain Monte Carlo (MCMC) and variational Bayes (VB), which are used to perform Bayesian inference using the posterior distribution. Simulations conducted on both synthetic and real datasets illustrate the benefits of the proposed methods in terms quality of deconvolution of the sparse samples.

Original languageEnglish
Title of host publication2018 IEEE Statistical Signal Processing Workshop (SSP)
PublisherIEEE
Pages603-607
Number of pages5
ISBN (Print)9781538615713
DOIs
Publication statusPublished - 30 Aug 2018
Event20th IEEE Statistical Signal Processing Workshop 2018 - Freiburg im Breisgau, Germany
Duration: 10 Jun 201813 Jun 2018

Conference

Conference20th IEEE Statistical Signal Processing Workshop 2018
Abbreviated titleSSP 2018
CountryGermany
CityFreiburg im Breisgau
Period10/06/1813/06/18

Fingerprint

Deconvolution
Additive noise
Convolution
Markov processes
Optical microscopy
Mathematical operators
Markov chains
inference
Imaging techniques
convolution integrals
bundles
field of view
microscopy
operators
simulation

Keywords

  • Bayesian estimation
  • Deconvolution
  • Irregular spatial sampling
  • Markov chain Monte Carlo methods
  • Variational Bayes

ASJC Scopus subject areas

  • Signal Processing
  • Instrumentation
  • Computer Networks and Communications

Cite this

Karam-Eldaly, A., Altmann, Y., Perperidis, A., & McLaughlin, S. (2018). Deconvolution of Irregularly Subsampled Images. In 2018 IEEE Statistical Signal Processing Workshop (SSP) (pp. 603-607). IEEE. https://doi.org/10.1109/SSP.2018.8450801
Karam-Eldaly, Ahmed ; Altmann, Yoann ; Perperidis, Antonios ; McLaughlin, Stephen. / Deconvolution of Irregularly Subsampled Images. 2018 IEEE Statistical Signal Processing Workshop (SSP). IEEE, 2018. pp. 603-607
@inproceedings{f9b8786ce58a4e47835b5bcd7b8e4e6d,
title = "Deconvolution of Irregularly Subsampled Images",
abstract = "Imaging technologies, such as coherent fibred bundle optical microscopy (FBOM), operate with irregularly-spaced sparse subsamples from their field of view. In this paper, we address the problem of data deconvolution for applications where the observed irregularly distributed samples are considered as a result of a convolution operator acting on original samples and corrupted by additive observation noise. We propose a hierarchical Bayesian model in which suitable prior distributions are assigned to the unknown model parameters, and compare two estimation strategies including Markov chain Monte Carlo (MCMC) and variational Bayes (VB), which are used to perform Bayesian inference using the posterior distribution. Simulations conducted on both synthetic and real datasets illustrate the benefits of the proposed methods in terms quality of deconvolution of the sparse samples.",
keywords = "Bayesian estimation, Deconvolution, Irregular spatial sampling, Markov chain Monte Carlo methods, Variational Bayes",
author = "Ahmed Karam-Eldaly and Yoann Altmann and Antonios Perperidis and Stephen McLaughlin",
year = "2018",
month = "8",
day = "30",
doi = "10.1109/SSP.2018.8450801",
language = "English",
isbn = "9781538615713",
pages = "603--607",
booktitle = "2018 IEEE Statistical Signal Processing Workshop (SSP)",
publisher = "IEEE",
address = "United States",

}

Karam-Eldaly, A, Altmann, Y, Perperidis, A & McLaughlin, S 2018, Deconvolution of Irregularly Subsampled Images. in 2018 IEEE Statistical Signal Processing Workshop (SSP). IEEE, pp. 603-607, 20th IEEE Statistical Signal Processing Workshop 2018, Freiburg im Breisgau, Germany, 10/06/18. https://doi.org/10.1109/SSP.2018.8450801

Deconvolution of Irregularly Subsampled Images. / Karam-Eldaly, Ahmed; Altmann, Yoann; Perperidis, Antonios; McLaughlin, Stephen.

2018 IEEE Statistical Signal Processing Workshop (SSP). IEEE, 2018. p. 603-607.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Deconvolution of Irregularly Subsampled Images

AU - Karam-Eldaly, Ahmed

AU - Altmann, Yoann

AU - Perperidis, Antonios

AU - McLaughlin, Stephen

PY - 2018/8/30

Y1 - 2018/8/30

N2 - Imaging technologies, such as coherent fibred bundle optical microscopy (FBOM), operate with irregularly-spaced sparse subsamples from their field of view. In this paper, we address the problem of data deconvolution for applications where the observed irregularly distributed samples are considered as a result of a convolution operator acting on original samples and corrupted by additive observation noise. We propose a hierarchical Bayesian model in which suitable prior distributions are assigned to the unknown model parameters, and compare two estimation strategies including Markov chain Monte Carlo (MCMC) and variational Bayes (VB), which are used to perform Bayesian inference using the posterior distribution. Simulations conducted on both synthetic and real datasets illustrate the benefits of the proposed methods in terms quality of deconvolution of the sparse samples.

AB - Imaging technologies, such as coherent fibred bundle optical microscopy (FBOM), operate with irregularly-spaced sparse subsamples from their field of view. In this paper, we address the problem of data deconvolution for applications where the observed irregularly distributed samples are considered as a result of a convolution operator acting on original samples and corrupted by additive observation noise. We propose a hierarchical Bayesian model in which suitable prior distributions are assigned to the unknown model parameters, and compare two estimation strategies including Markov chain Monte Carlo (MCMC) and variational Bayes (VB), which are used to perform Bayesian inference using the posterior distribution. Simulations conducted on both synthetic and real datasets illustrate the benefits of the proposed methods in terms quality of deconvolution of the sparse samples.

KW - Bayesian estimation

KW - Deconvolution

KW - Irregular spatial sampling

KW - Markov chain Monte Carlo methods

KW - Variational Bayes

UR - http://www.scopus.com/inward/record.url?scp=85053831839&partnerID=8YFLogxK

U2 - 10.1109/SSP.2018.8450801

DO - 10.1109/SSP.2018.8450801

M3 - Conference contribution

SN - 9781538615713

SP - 603

EP - 607

BT - 2018 IEEE Statistical Signal Processing Workshop (SSP)

PB - IEEE

ER -

Karam-Eldaly A, Altmann Y, Perperidis A, McLaughlin S. Deconvolution of Irregularly Subsampled Images. In 2018 IEEE Statistical Signal Processing Workshop (SSP). IEEE. 2018. p. 603-607 https://doi.org/10.1109/SSP.2018.8450801