A contrario selection of optimal partitions for image segmentation

We present a novel segmentation algorithm based on a hierarchical representation of images. The main contribution of this work is to explore the capabilities of the a contrario reasoning when applied to the segmentation problem and to overcome the limitations of current algorithms within that framew...

Full description

Bibliographic Details
Authors: Cardelino, Juan, Caselles, Vicente, Bertalmío, Marcelo, Randall, Gregory
Format: article
Status:Published version
Publication Date:2013
Country:España
Institution:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repository:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:recercat.cat:10230/26982
Online Access:http://hdl.handle.net/10230/26982
http://dx.doi.org/10.1137/11086029X
Access Level:Open access
Keyword:A contrario methods
Segmentation
Hierarchy
Quantitative evaluation
id ES_dcd96ca3ce201e9a7b6d3fba815294d8
oai_identifier_str oai:recercat.cat:10230/26982
network_acronym_str ES
network_name_str España
repository_id_str
spelling A contrario selection of optimal partitions for image segmentationCardelino, JuanCaselles, VicenteBertalmío, MarceloRandall, GregoryA contrario methodsSegmentationHierarchyQuantitative evaluationWe present a novel segmentation algorithm based on a hierarchical representation of images. The main contribution of this work is to explore the capabilities of the a contrario reasoning when applied to the segmentation problem and to overcome the limitations of current algorithms within that framework. This exploratory approach has three main goals. Our first goal is to extend the search space of greedy merging algorithms to the set of all partitions spanned by a certain hierarchy and to cast the segmentation as a selection problem within this space. In this way we increase the number of tested partitions, and thus we potentially improve the segmentation results. In addition, this space is considerably smaller than the space of all possible partitions, and thus we still keep the complexity controlled. Our second goal aims to improve the locality of region merging algorithms, which usually merge pairs of neighboring regions. In this work, we overcome this limitation by introducing a validation procedure for complete partitions rather than for pairs of regions. The third goal is to perform an exhaustive experimental evaluation methodology in order to provide reproducible results. Finally, we embed the selection process on a statistical a contrario framework which allows us to have only one free parameter related to the desired scale.J. Cardelino and V. Caselles acknowledge partial support by MICINN project, reference MTM2009-08171 and by GRC, reference 2009 SGR 773, funded by the Generalitat de Catalunya. V. Caselles also acknowledges partial support by ”ICREA Acade`mia” prize for excellence in research funded by the Generalitat de Catalunya, and by the ERC Advanced Grant INPAINTING (Grant agreement no.: 319899). M. Bertalm´ıo acknowledges support by European Research Council, Starting Grant ref. 306337, and Spanish grants AACC, ref. TIN2011-15954-E, and Plan Nacional, ref. TIN2012-38112. J. Cardelino also acknowledges partial support by ALFA-CVFA project and Tecnocom scolarship.SIAM (Society for Industrial and Applied Mathematics)201620162013info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttp://hdl.handle.net/10230/26982http://dx.doi.org/10.1137/11086029Xreponame:Recercat. Dipósit de la Recerca de Catalunyainstname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)InglésSIAM Journal on Imaging Sciences. 2013;6(3):1274-317.info:eu-repo/grantAgreement/EC/FP7/306337© Society for Industrial and Applied Mathematicsinfo:eu-repo/semantics/openAccessoai:recercat.cat:10230/269822026-05-29T05:05:01Z
dc.title.none.fl_str_mv A contrario selection of optimal partitions for image segmentation
title A contrario selection of optimal partitions for image segmentation
spellingShingle A contrario selection of optimal partitions for image segmentation
Cardelino, Juan
A contrario methods
Segmentation
Hierarchy
Quantitative evaluation
title_short A contrario selection of optimal partitions for image segmentation
title_full A contrario selection of optimal partitions for image segmentation
title_fullStr A contrario selection of optimal partitions for image segmentation
title_full_unstemmed A contrario selection of optimal partitions for image segmentation
title_sort A contrario selection of optimal partitions for image segmentation
dc.creator.none.fl_str_mv Cardelino, Juan
Caselles, Vicente
Bertalmío, Marcelo
Randall, Gregory
author Cardelino, Juan
author_facet Cardelino, Juan
Caselles, Vicente
Bertalmío, Marcelo
Randall, Gregory
author_role author
author2 Caselles, Vicente
Bertalmío, Marcelo
Randall, Gregory
author2_role author
author
author
dc.subject.none.fl_str_mv A contrario methods
Segmentation
Hierarchy
Quantitative evaluation
topic A contrario methods
Segmentation
Hierarchy
Quantitative evaluation
description We present a novel segmentation algorithm based on a hierarchical representation of images. The main contribution of this work is to explore the capabilities of the a contrario reasoning when applied to the segmentation problem and to overcome the limitations of current algorithms within that framework. This exploratory approach has three main goals. Our first goal is to extend the search space of greedy merging algorithms to the set of all partitions spanned by a certain hierarchy and to cast the segmentation as a selection problem within this space. In this way we increase the number of tested partitions, and thus we potentially improve the segmentation results. In addition, this space is considerably smaller than the space of all possible partitions, and thus we still keep the complexity controlled. Our second goal aims to improve the locality of region merging algorithms, which usually merge pairs of neighboring regions. In this work, we overcome this limitation by introducing a validation procedure for complete partitions rather than for pairs of regions. The third goal is to perform an exhaustive experimental evaluation methodology in order to provide reproducible results. Finally, we embed the selection process on a statistical a contrario framework which allows us to have only one free parameter related to the desired scale.
publishDate 2013
dc.date.none.fl_str_mv 2013
2016
2016
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv http://hdl.handle.net/10230/26982
http://dx.doi.org/10.1137/11086029X
url http://hdl.handle.net/10230/26982
http://dx.doi.org/10.1137/11086029X
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv SIAM Journal on Imaging Sciences. 2013;6(3):1274-317.
info:eu-repo/grantAgreement/EC/FP7/306337
dc.rights.none.fl_str_mv © Society for Industrial and Applied Mathematics
info:eu-repo/semantics/openAccess
rights_invalid_str_mv © Society for Industrial and Applied Mathematics
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv SIAM (Society for Industrial and Applied Mathematics)
publisher.none.fl_str_mv SIAM (Society for Industrial and Applied Mathematics)
dc.source.none.fl_str_mv reponame:Recercat. Dipósit de la Recerca de Catalunya
instname:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
instname_str Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
reponame_str Recercat. Dipósit de la Recerca de Catalunya
collection Recercat. Dipósit de la Recerca de Catalunya
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869421806481309696
score 15,812429

Similar Items