File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Unsupervised segmentation of natural images via lossy data compression

TitleUnsupervised segmentation of natural images via lossy data compression
Authors
KeywordsClustering
Image segmentation
Lossy compression
Mixture of Gaussian distributions
Texture segmentation
Issue Date2008
Citation
Computer Vision and Image Understanding, 2008, v. 110, n. 2, p. 212-225 How to Cite?
AbstractIn this paper, we cast natural-image segmentation as a problem of clustering texture features as multivariate mixed data. We model the distribution of the texture features using a mixture of Gaussian distributions. Unlike most existing clustering methods, we allow the mixture components to be degenerate or nearly-degenerate. We contend that this assumption is particularly important for mid-level image segmentation, where degeneracy is typically introduced by using a common feature representation for different textures in an image. We show that such a mixture distribution can be effectively segmented by a simple agglomerative clustering algorithm derived from a lossy data compression approach. Using either 2D texture filter banks or simple fixed-size windows to obtain texture features, the algorithm effectively segments an image by minimizing the overall coding length of the feature vectors. We conduct comprehensive experiments to measure the performance of the algorithm in terms of visual evaluation and a variety of quantitative indices for image segmentation. The algorithm compares favorably against other well-known image-segmentation methods on the Berkeley image database. © 2007 Elsevier Inc. All rights reserved.
Persistent Identifierhttp://hdl.handle.net/10722/326740
ISSN
2023 Impact Factor: 4.3
2023 SCImago Journal Rankings: 1.420
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorYang, Allen Y.-
dc.contributor.authorWright, John-
dc.contributor.authorMa, Yi-
dc.contributor.authorSastry, S. Shankar-
dc.date.accessioned2023-03-31T05:26:11Z-
dc.date.available2023-03-31T05:26:11Z-
dc.date.issued2008-
dc.identifier.citationComputer Vision and Image Understanding, 2008, v. 110, n. 2, p. 212-225-
dc.identifier.issn1077-3142-
dc.identifier.urihttp://hdl.handle.net/10722/326740-
dc.description.abstractIn this paper, we cast natural-image segmentation as a problem of clustering texture features as multivariate mixed data. We model the distribution of the texture features using a mixture of Gaussian distributions. Unlike most existing clustering methods, we allow the mixture components to be degenerate or nearly-degenerate. We contend that this assumption is particularly important for mid-level image segmentation, where degeneracy is typically introduced by using a common feature representation for different textures in an image. We show that such a mixture distribution can be effectively segmented by a simple agglomerative clustering algorithm derived from a lossy data compression approach. Using either 2D texture filter banks or simple fixed-size windows to obtain texture features, the algorithm effectively segments an image by minimizing the overall coding length of the feature vectors. We conduct comprehensive experiments to measure the performance of the algorithm in terms of visual evaluation and a variety of quantitative indices for image segmentation. The algorithm compares favorably against other well-known image-segmentation methods on the Berkeley image database. © 2007 Elsevier Inc. All rights reserved.-
dc.languageeng-
dc.relation.ispartofComputer Vision and Image Understanding-
dc.subjectClustering-
dc.subjectImage segmentation-
dc.subjectLossy compression-
dc.subjectMixture of Gaussian distributions-
dc.subjectTexture segmentation-
dc.titleUnsupervised segmentation of natural images via lossy data compression-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1016/j.cviu.2007.07.005-
dc.identifier.scopuseid_2-s2.0-41949137770-
dc.identifier.volume110-
dc.identifier.issue2-
dc.identifier.spage212-
dc.identifier.epage225-
dc.identifier.eissn1090-235X-
dc.identifier.isiWOS:000255322900004-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats