Robust image registration using adaptive expectation maximisation based PCA

Parminder Singh Reel, Laurence S. Dooley, K. C. P. Wong, Anko Borner

Research output: Contribution to conferencePaper

Abstract

Images having either the same or different modalities can be aligned using the systematic process of image registration. Inherent image characteristics including intensity non-uniformities in magnetic resonance images and large homogeneous non-vascular regions in retinal and other generic image types however, pose a significant challenge to their registration. This paper presents an adaptive expectation maximisation for principal component analysis with mutual information (aEMPCA-MI) similarity measure for image registration. It introduces a novel iterative process to adaptively select the most significant principal components using Kaiser rule and applies 4-pixel connectivity for feature extraction together with Wichard's bin size selection in calculating the MI. Both quantitative and qualitative results on a diverse range of image datasets, conclusively demonstrate the superior image registration performance of aEMPCA-MI compared with existing Mi-based similarity measures.
Original languageEnglish
Pages105-108
Number of pages4
DOIs
Publication statusPublished - Dec 2014
Event2014 Visual Communications and Image Processing (VCIP) - Valletta, Malta
Duration: 7 Dec 201410 Dec 2014

Conference

Conference2014 Visual Communications and Image Processing (VCIP)
Period7/12/1410/12/14

Fingerprint

Image registration
Principal component analysis
Bins
Magnetic resonance
Feature extraction
Pixels

Keywords

  • Retina
  • Image registration
  • Principal component analysis
  • Mutual information
  • Subspace constraints
  • Magnetic resonance imaging
  • Robustness

Cite this

Reel, P. S., Dooley, L. S., Wong, K. C. P., & Borner, A. (2014). Robust image registration using adaptive expectation maximisation based PCA. 105-108. Paper presented at 2014 Visual Communications and Image Processing (VCIP), . https://doi.org/10.1109/VCIP.2014.7051515
Reel, Parminder Singh ; Dooley, Laurence S. ; Wong, K. C. P. ; Borner, Anko. / Robust image registration using adaptive expectation maximisation based PCA. Paper presented at 2014 Visual Communications and Image Processing (VCIP), .4 p.
@conference{3177538c5dfa4cb38e30fb0d9c01c81f,
title = "Robust image registration using adaptive expectation maximisation based PCA",
abstract = "Images having either the same or different modalities can be aligned using the systematic process of image registration. Inherent image characteristics including intensity non-uniformities in magnetic resonance images and large homogeneous non-vascular regions in retinal and other generic image types however, pose a significant challenge to their registration. This paper presents an adaptive expectation maximisation for principal component analysis with mutual information (aEMPCA-MI) similarity measure for image registration. It introduces a novel iterative process to adaptively select the most significant principal components using Kaiser rule and applies 4-pixel connectivity for feature extraction together with Wichard's bin size selection in calculating the MI. Both quantitative and qualitative results on a diverse range of image datasets, conclusively demonstrate the superior image registration performance of aEMPCA-MI compared with existing Mi-based similarity measures.",
keywords = "Retina, Image registration, Principal component analysis, Mutual information, Subspace constraints, Magnetic resonance imaging, Robustness",
author = "Reel, {Parminder Singh} and Dooley, {Laurence S.} and Wong, {K. C. P.} and Anko Borner",
year = "2014",
month = "12",
doi = "10.1109/VCIP.2014.7051515",
language = "English",
pages = "105--108",
note = "2014 Visual Communications and Image Processing (VCIP) ; Conference date: 07-12-2014 Through 10-12-2014",

}

Reel, PS, Dooley, LS, Wong, KCP & Borner, A 2014, 'Robust image registration using adaptive expectation maximisation based PCA' Paper presented at 2014 Visual Communications and Image Processing (VCIP), 7/12/14 - 10/12/14, pp. 105-108. https://doi.org/10.1109/VCIP.2014.7051515

Robust image registration using adaptive expectation maximisation based PCA. / Reel, Parminder Singh; Dooley, Laurence S.; Wong, K. C. P.; Borner, Anko.

2014. 105-108 Paper presented at 2014 Visual Communications and Image Processing (VCIP), .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Robust image registration using adaptive expectation maximisation based PCA

AU - Reel, Parminder Singh

AU - Dooley, Laurence S.

AU - Wong, K. C. P.

AU - Borner, Anko

PY - 2014/12

Y1 - 2014/12

N2 - Images having either the same or different modalities can be aligned using the systematic process of image registration. Inherent image characteristics including intensity non-uniformities in magnetic resonance images and large homogeneous non-vascular regions in retinal and other generic image types however, pose a significant challenge to their registration. This paper presents an adaptive expectation maximisation for principal component analysis with mutual information (aEMPCA-MI) similarity measure for image registration. It introduces a novel iterative process to adaptively select the most significant principal components using Kaiser rule and applies 4-pixel connectivity for feature extraction together with Wichard's bin size selection in calculating the MI. Both quantitative and qualitative results on a diverse range of image datasets, conclusively demonstrate the superior image registration performance of aEMPCA-MI compared with existing Mi-based similarity measures.

AB - Images having either the same or different modalities can be aligned using the systematic process of image registration. Inherent image characteristics including intensity non-uniformities in magnetic resonance images and large homogeneous non-vascular regions in retinal and other generic image types however, pose a significant challenge to their registration. This paper presents an adaptive expectation maximisation for principal component analysis with mutual information (aEMPCA-MI) similarity measure for image registration. It introduces a novel iterative process to adaptively select the most significant principal components using Kaiser rule and applies 4-pixel connectivity for feature extraction together with Wichard's bin size selection in calculating the MI. Both quantitative and qualitative results on a diverse range of image datasets, conclusively demonstrate the superior image registration performance of aEMPCA-MI compared with existing Mi-based similarity measures.

KW - Retina

KW - Image registration

KW - Principal component analysis

KW - Mutual information

KW - Subspace constraints

KW - Magnetic resonance imaging

KW - Robustness

U2 - 10.1109/VCIP.2014.7051515

DO - 10.1109/VCIP.2014.7051515

M3 - Paper

SP - 105

EP - 108

ER -

Reel PS, Dooley LS, Wong KCP, Borner A. Robust image registration using adaptive expectation maximisation based PCA. 2014. Paper presented at 2014 Visual Communications and Image Processing (VCIP), . https://doi.org/10.1109/VCIP.2014.7051515