A Noise Robust Statistical Texture Model

Klaus Baggesen Hilger; Mikkel Bille Stegmann; Rasmus Larsen; T. Dohi

A Noise Robust Statistical Texture Model

Klaus Baggesen Hilger, Mikkel Bille Stegmann, Rasmus Larsen, T. Dohi (Editor)

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

138 Downloads (Pure)

Abstract

This paper presents a novel approach to the problem of obtaining a low dimensional representation of texture (pixel intensity) variation present in a training set after alignment using a Generalised Procrustes analysis.We extend the conventional analysis of training textures in the Active Appearance Models segmentation framework. This is accomplished by augmenting the model with an estimate of the covariance of the noise present in the training data. This results in a more compact model maximising the signal-to-noise ratio, thus favouring subspaces rich on signal, but low on noise. Differences in the methods are illustrated on a set of left cardiac ventricles obtained using magnetic resonance imaging.

Original language	English
Title of host publication	Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, Tokyo, Japan
Publisher	Springer
Publication date	2002
Pages	444-451
Publication status	Published - 2002
Event	Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, - Tokyo, Japan Duration: 1 Jan 2002 → …

Conference

Conference	Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference,
City	Tokyo, Japan
Period	01/01/2002 → …

Access to Document

Imm856Final published version, 340 KB

Fingerprint Dive into the research topics of 'A Noise Robust Statistical Texture Model'. Together they form a unique fingerprint.

Cite this

@inproceedings{623aeafb64e74d9a833d9430daea199c,

title = "A Noise Robust Statistical Texture Model",

abstract = "This paper presents a novel approach to the problem of obtaining a low dimensional representation of texture (pixel intensity) variation present in a training set after alignment using a Generalised Procrustes analysis.We extend the conventional analysis of training textures in the Active Appearance Models segmentation framework. This is accomplished by augmenting the model with an estimate of the covariance of the noise present in the training data. This results in a more compact model maximising the signal-to-noise ratio, thus favouring subspaces rich on signal, but low on noise. Differences in the methods are illustrated on a set of left cardiac ventricles obtained using magnetic resonance imaging.",

author = "Hilger, {Klaus Baggesen} and Stegmann, {Mikkel Bille} and Rasmus Larsen and T. Dohi",

year = "2002",

language = "English",

pages = "444--451",

booktitle = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, Tokyo, Japan",

publisher = "Springer",

note = "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, ; Conference date: 01-01-2002",

}

A Noise Robust Statistical Texture Model. / Hilger, Klaus Baggesen; Stegmann, Mikkel Bille; Larsen, Rasmus; Dohi, T. (Editor).

Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, Tokyo, Japan. Springer, 2002. p. 444-451.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - A Noise Robust Statistical Texture Model

AU - Hilger, Klaus Baggesen

AU - Stegmann, Mikkel Bille

AU - Larsen, Rasmus

A2 - Dohi, T.

PY - 2002

Y1 - 2002

N2 - This paper presents a novel approach to the problem of obtaining a low dimensional representation of texture (pixel intensity) variation present in a training set after alignment using a Generalised Procrustes analysis.We extend the conventional analysis of training textures in the Active Appearance Models segmentation framework. This is accomplished by augmenting the model with an estimate of the covariance of the noise present in the training data. This results in a more compact model maximising the signal-to-noise ratio, thus favouring subspaces rich on signal, but low on noise. Differences in the methods are illustrated on a set of left cardiac ventricles obtained using magnetic resonance imaging.

AB - This paper presents a novel approach to the problem of obtaining a low dimensional representation of texture (pixel intensity) variation present in a training set after alignment using a Generalised Procrustes analysis.We extend the conventional analysis of training textures in the Active Appearance Models segmentation framework. This is accomplished by augmenting the model with an estimate of the covariance of the noise present in the training data. This results in a more compact model maximising the signal-to-noise ratio, thus favouring subspaces rich on signal, but low on noise. Differences in the methods are illustrated on a set of left cardiac ventricles obtained using magnetic resonance imaging.

M3 - Article in proceedings

SP - 444

EP - 451

BT - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference, Tokyo, Japan

PB - Springer

T2 - Medical Image Computing and Computer-Assisted Intervention - MICCAI 2002, 5th Int. Conference,

Y2 - 1 January 2002

ER -