Profile scale-spaces for multiscale image match

Sean Ho; Guido Gerig

Profile scale-spaces for multiscale image match

Sean Ho, Guido Gerig

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

Abstract

We present a novel statistical image-match model for use in Bayesian segmentation, a multiscale extension of image profile models akin to those in Active Shape Models. A spherical-harmonic based 3D shape representation provides a mapping of the object boundary to the sphere S², and a scale-space for profiles on the sphere defines a scale-space on the object. A key feature is that profiles are not blurred across the object boundary, but only along the boundary. This profile scalespace is sampled in a coarse-to-fine fashion to produce features for the statistical image-match model. A framework for model-building and segmentation has been built, and testing and validation are in progress with a dataset of 70 segmented images of the caudate nucleus.

Original language	English (US)
Title of host publication	Lecture Notes in Computer Science
Editors	C. Barillot, D.R. Haynor, P. Hellier
Pages	176-183
Number of pages	8
Volume	3216
Edition	PART 1
State	Published - 2004
Event	Medical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings - Saint-Malo, France Duration: Sep 26 2004 → Sep 29 2004

Other

Other	Medical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings
Country/Territory	France
City	Saint-Malo
Period	9/26/04 → 9/29/04

ASJC Scopus subject areas

Computer Science (miscellaneous)
General Computer Science
Theoretical Computer Science

Cite this

@inproceedings{bdde9a4e16994923baca5938e705faa7,

title = "Profile scale-spaces for multiscale image match",

abstract = "We present a novel statistical image-match model for use in Bayesian segmentation, a multiscale extension of image profile models akin to those in Active Shape Models. A spherical-harmonic based 3D shape representation provides a mapping of the object boundary to the sphere S2, and a scale-space for profiles on the sphere defines a scale-space on the object. A key feature is that profiles are not blurred across the object boundary, but only along the boundary. This profile scalespace is sampled in a coarse-to-fine fashion to produce features for the statistical image-match model. A framework for model-building and segmentation has been built, and testing and validation are in progress with a dataset of 70 segmented images of the caudate nucleus.",

author = "Sean Ho and Guido Gerig",

year = "2004",

language = "English (US)",

volume = "3216",

pages = "176--183",

editor = "C. Barillot and D.R. Haynor and P. Hellier",

booktitle = "Lecture Notes in Computer Science",

edition = "PART 1",

note = "Medical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings ; Conference date: 26-09-2004 Through 29-09-2004",

}

TY - GEN

T1 - Profile scale-spaces for multiscale image match

AU - Ho, Sean

AU - Gerig, Guido

PY - 2004

Y1 - 2004

N2 - We present a novel statistical image-match model for use in Bayesian segmentation, a multiscale extension of image profile models akin to those in Active Shape Models. A spherical-harmonic based 3D shape representation provides a mapping of the object boundary to the sphere S2, and a scale-space for profiles on the sphere defines a scale-space on the object. A key feature is that profiles are not blurred across the object boundary, but only along the boundary. This profile scalespace is sampled in a coarse-to-fine fashion to produce features for the statistical image-match model. A framework for model-building and segmentation has been built, and testing and validation are in progress with a dataset of 70 segmented images of the caudate nucleus.

AB - We present a novel statistical image-match model for use in Bayesian segmentation, a multiscale extension of image profile models akin to those in Active Shape Models. A spherical-harmonic based 3D shape representation provides a mapping of the object boundary to the sphere S2, and a scale-space for profiles on the sphere defines a scale-space on the object. A key feature is that profiles are not blurred across the object boundary, but only along the boundary. This profile scalespace is sampled in a coarse-to-fine fashion to produce features for the statistical image-match model. A framework for model-building and segmentation has been built, and testing and validation are in progress with a dataset of 70 segmented images of the caudate nucleus.

UR - http://www.scopus.com/inward/record.url?scp=20344374965&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=20344374965&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:20344374965

VL - 3216

SP - 176

EP - 183

BT - Lecture Notes in Computer Science

A2 - Barillot, C.

A2 - Haynor, D.R.

A2 - Hellier, P.

T2 - Medical Image Computing and Computer-Assisted Intervention, MICCAI 2004 - 7th International Conference, Proceedings

Y2 - 26 September 2004 through 29 September 2004

ER -

Profile scale-spaces for multiscale image match

Abstract

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this