Images as occlusions of textures: A framework for segmentation

Michael T. McCann, Dustin G. Mixon, Matthew C. Fickus, Carlos A. Castro, John A. Ozolek, Jelena Kovacevic

Research output: Contribution to journalArticle

Abstract

We propose a new mathematical and algorithmic framework for unsupervised image segmentation, which is a critical step in a wide variety of image processing applications. We have found that most existing segmentation methods are not successful on histopathology images, which prompted us to investigate segmentation of a broader class of images, namely those without clear edges between the regions to be segmented. We model these images as occlusions of random images, which we call textures, and show that local histograms are a useful tool for segmenting them. Based on our theoretical results, we describe a flexible segmentation framework that draws on existing work on nonnegative matrix factorization and image deconvolution. Results on synthetic texture mosaics and real histology images show the promise of the method.

Original languageEnglish (US)
Article number6748922
Pages (from-to)2033-2046
Number of pages14
JournalIEEE Transactions on Image Processing
Volume23
Issue number5
DOIs
StatePublished - Jan 1 2014

Fingerprint

Textures
Histology
Deconvolution
Factorization
Image segmentation
Image processing

Keywords

  • deconvolution
  • image segmentation
  • local histograms
  • non-negative matrix factorization
  • occlusion models
  • texture

ASJC Scopus subject areas

  • Software
  • Computer Graphics and Computer-Aided Design

Cite this

McCann, M. T., Mixon, D. G., Fickus, M. C., Castro, C. A., Ozolek, J. A., & Kovacevic, J. (2014). Images as occlusions of textures: A framework for segmentation. IEEE Transactions on Image Processing, 23(5), 2033-2046. [6748922]. https://doi.org/10.1109/TIP.2014.2307475

Images as occlusions of textures : A framework for segmentation. / McCann, Michael T.; Mixon, Dustin G.; Fickus, Matthew C.; Castro, Carlos A.; Ozolek, John A.; Kovacevic, Jelena.

In: IEEE Transactions on Image Processing, Vol. 23, No. 5, 6748922, 01.01.2014, p. 2033-2046.

Research output: Contribution to journalArticle

McCann, MT, Mixon, DG, Fickus, MC, Castro, CA, Ozolek, JA & Kovacevic, J 2014, 'Images as occlusions of textures: A framework for segmentation', IEEE Transactions on Image Processing, vol. 23, no. 5, 6748922, pp. 2033-2046. https://doi.org/10.1109/TIP.2014.2307475
McCann, Michael T. ; Mixon, Dustin G. ; Fickus, Matthew C. ; Castro, Carlos A. ; Ozolek, John A. ; Kovacevic, Jelena. / Images as occlusions of textures : A framework for segmentation. In: IEEE Transactions on Image Processing. 2014 ; Vol. 23, No. 5. pp. 2033-2046.
@article{f23f40428c694692887124562f2c3527,
title = "Images as occlusions of textures: A framework for segmentation",
abstract = "We propose a new mathematical and algorithmic framework for unsupervised image segmentation, which is a critical step in a wide variety of image processing applications. We have found that most existing segmentation methods are not successful on histopathology images, which prompted us to investigate segmentation of a broader class of images, namely those without clear edges between the regions to be segmented. We model these images as occlusions of random images, which we call textures, and show that local histograms are a useful tool for segmenting them. Based on our theoretical results, we describe a flexible segmentation framework that draws on existing work on nonnegative matrix factorization and image deconvolution. Results on synthetic texture mosaics and real histology images show the promise of the method.",
keywords = "deconvolution, image segmentation, local histograms, non-negative matrix factorization, occlusion models, texture",
author = "McCann, {Michael T.} and Mixon, {Dustin G.} and Fickus, {Matthew C.} and Castro, {Carlos A.} and Ozolek, {John A.} and Jelena Kovacevic",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/TIP.2014.2307475",
language = "English (US)",
volume = "23",
pages = "2033--2046",
journal = "IEEE Transactions on Image Processing",
issn = "1057-7149",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Images as occlusions of textures

T2 - A framework for segmentation

AU - McCann, Michael T.

AU - Mixon, Dustin G.

AU - Fickus, Matthew C.

AU - Castro, Carlos A.

AU - Ozolek, John A.

AU - Kovacevic, Jelena

PY - 2014/1/1

Y1 - 2014/1/1

N2 - We propose a new mathematical and algorithmic framework for unsupervised image segmentation, which is a critical step in a wide variety of image processing applications. We have found that most existing segmentation methods are not successful on histopathology images, which prompted us to investigate segmentation of a broader class of images, namely those without clear edges between the regions to be segmented. We model these images as occlusions of random images, which we call textures, and show that local histograms are a useful tool for segmenting them. Based on our theoretical results, we describe a flexible segmentation framework that draws on existing work on nonnegative matrix factorization and image deconvolution. Results on synthetic texture mosaics and real histology images show the promise of the method.

AB - We propose a new mathematical and algorithmic framework for unsupervised image segmentation, which is a critical step in a wide variety of image processing applications. We have found that most existing segmentation methods are not successful on histopathology images, which prompted us to investigate segmentation of a broader class of images, namely those without clear edges between the regions to be segmented. We model these images as occlusions of random images, which we call textures, and show that local histograms are a useful tool for segmenting them. Based on our theoretical results, we describe a flexible segmentation framework that draws on existing work on nonnegative matrix factorization and image deconvolution. Results on synthetic texture mosaics and real histology images show the promise of the method.

KW - deconvolution

KW - image segmentation

KW - local histograms

KW - non-negative matrix factorization

KW - occlusion models

KW - texture

UR - http://www.scopus.com/inward/record.url?scp=84898037550&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898037550&partnerID=8YFLogxK

U2 - 10.1109/TIP.2014.2307475

DO - 10.1109/TIP.2014.2307475

M3 - Article

C2 - 24710403

AN - SCOPUS:84898037550

VL - 23

SP - 2033

EP - 2046

JO - IEEE Transactions on Image Processing

JF - IEEE Transactions on Image Processing

SN - 1057-7149

IS - 5

M1 - 6748922

ER -