Non-rigid registration using gradient of self-similarity response

James L. Huang, Jeffrey J Rodriguez

Research output: Contribution to journalArticle

Abstract

Locally affine transformation with globally elastic interpolation is a common strategy for non-rigid registration. Current techniques improve the registration accuracy by only processing the sub-images that contain well-defined structures quantified by Moran's spatial correlation. As an indicator, Moran's metric successfully excludes noisy structures that result in misleading global optimum in terms of similarity. However, some well-defined structures with intensity only varying in one direction may also cause mis-registration. In this paper, we propose a new metric based on the response of a similarity function to quantify the ability of being correctly registered for each sub-image. Using receiver operating characteristic analysis, we show that the proposed metric more accurately reflects such ability than Moran's metric. Incorporating the proposed metric into a hierarchical non-rigid registration scheme, we show that registration accuracy is improved relative to Moran's metric.

Original languageEnglish (US)
Pages (from-to)825-834
Number of pages10
JournalImage and Vision Computing
Volume32
Issue number11
DOIs
StatePublished - 2014

Fingerprint

Interpolation
Processing

Keywords

  • Hierarchical elastic registration
  • Locally affine transformation
  • Non-rigid registration

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

Non-rigid registration using gradient of self-similarity response. / Huang, James L.; Rodriguez, Jeffrey J.

In: Image and Vision Computing, Vol. 32, No. 11, 2014, p. 825-834.

Research output: Contribution to journalArticle

@article{c86de891623c411da1c2ccfa2e5f0125,
title = "Non-rigid registration using gradient of self-similarity response",
abstract = "Locally affine transformation with globally elastic interpolation is a common strategy for non-rigid registration. Current techniques improve the registration accuracy by only processing the sub-images that contain well-defined structures quantified by Moran's spatial correlation. As an indicator, Moran's metric successfully excludes noisy structures that result in misleading global optimum in terms of similarity. However, some well-defined structures with intensity only varying in one direction may also cause mis-registration. In this paper, we propose a new metric based on the response of a similarity function to quantify the ability of being correctly registered for each sub-image. Using receiver operating characteristic analysis, we show that the proposed metric more accurately reflects such ability than Moran's metric. Incorporating the proposed metric into a hierarchical non-rigid registration scheme, we show that registration accuracy is improved relative to Moran's metric.",
keywords = "Hierarchical elastic registration, Locally affine transformation, Non-rigid registration",
author = "Huang, {James L.} and Rodriguez, {Jeffrey J}",
year = "2014",
doi = "10.1016/j.imavis.2014.06.005",
language = "English (US)",
volume = "32",
pages = "825--834",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier Limited",
number = "11",

}

TY - JOUR

T1 - Non-rigid registration using gradient of self-similarity response

AU - Huang, James L.

AU - Rodriguez, Jeffrey J

PY - 2014

Y1 - 2014

N2 - Locally affine transformation with globally elastic interpolation is a common strategy for non-rigid registration. Current techniques improve the registration accuracy by only processing the sub-images that contain well-defined structures quantified by Moran's spatial correlation. As an indicator, Moran's metric successfully excludes noisy structures that result in misleading global optimum in terms of similarity. However, some well-defined structures with intensity only varying in one direction may also cause mis-registration. In this paper, we propose a new metric based on the response of a similarity function to quantify the ability of being correctly registered for each sub-image. Using receiver operating characteristic analysis, we show that the proposed metric more accurately reflects such ability than Moran's metric. Incorporating the proposed metric into a hierarchical non-rigid registration scheme, we show that registration accuracy is improved relative to Moran's metric.

AB - Locally affine transformation with globally elastic interpolation is a common strategy for non-rigid registration. Current techniques improve the registration accuracy by only processing the sub-images that contain well-defined structures quantified by Moran's spatial correlation. As an indicator, Moran's metric successfully excludes noisy structures that result in misleading global optimum in terms of similarity. However, some well-defined structures with intensity only varying in one direction may also cause mis-registration. In this paper, we propose a new metric based on the response of a similarity function to quantify the ability of being correctly registered for each sub-image. Using receiver operating characteristic analysis, we show that the proposed metric more accurately reflects such ability than Moran's metric. Incorporating the proposed metric into a hierarchical non-rigid registration scheme, we show that registration accuracy is improved relative to Moran's metric.

KW - Hierarchical elastic registration

KW - Locally affine transformation

KW - Non-rigid registration

UR - http://www.scopus.com/inward/record.url?scp=84906564702&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906564702&partnerID=8YFLogxK

U2 - 10.1016/j.imavis.2014.06.005

DO - 10.1016/j.imavis.2014.06.005

M3 - Article

AN - SCOPUS:84906564702

VL - 32

SP - 825

EP - 834

JO - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

IS - 11

ER -