Loading…

Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation

We generalize the method of Slow Feature Analysis (SFA) for vector-valued functions of several variables and apply it to the problem of blind source separation, in particular to image separation. It is generally necessary to use multivariate SFA instead of univariate SFA for separating multi-dimensi...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on image processing 2013-07, Vol.22 (7), p.2737-2750
Main Authors: Ha Quang Minh, Wiskott, Laurenz
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by cdi_FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803
cites cdi_FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803
container_end_page 2750
container_issue 7
container_start_page 2737
container_title IEEE transactions on image processing
container_volume 22
creator Ha Quang Minh
Wiskott, Laurenz
description We generalize the method of Slow Feature Analysis (SFA) for vector-valued functions of several variables and apply it to the problem of blind source separation, in particular to image separation. It is generally necessary to use multivariate SFA instead of univariate SFA for separating multi-dimensional signals. For the linear case, an exact mathematical analysis is given, which shows in particular that the sources are perfectly separated by SFA if and only if they and their first-order derivatives are uncorrelated. When the sources are correlated, we apply the following technique called Decorrelation Filtering: use a linear filter to decorrelate the sources and their derivatives in the given mixture, then apply the unmixing matrix obtained on the filtered mixtures to the original mixtures. If the filtered sources are perfectly separated by this matrix, so are the original sources. A decorrelation filter can be numerically obtained by solving a nonlinear optimization problem. This technique can also be applied to other linear separation methods, whose output signals are decorrelated, such as ICA. When there are more mixtures than sources, one can determine the actual number of sources by using a regularized version of SFA with decorrelation filtering. Extensive numerical experiments using SFA and ICA with decorrelation filtering, supported by mathematical analysis, demonstrate the potential of our methods for solving problems involving blind source separation.
doi_str_mv 10.1109/TIP.2013.2257808
format article
fullrecord <record><control><sourceid>proquest_pasca</sourceid><recordid>TN_cdi_pascalfrancis_primary_27599764</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6497610</ieee_id><sourcerecordid>1353477088</sourcerecordid><originalsourceid>FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803</originalsourceid><addsrcrecordid>eNpFkE1LxDAQhoMoft8FQXoRvHSdadOmOfq1KigK7slLSdOJRLLtmrSK_97ornqagXneF-Zh7ABhggjydHb7OMkA80mWFaKCao1to-SYAvBsPe5QiFQgl1tsJ4RXAOQFlptsK8sLibyS2-z5fnSDfVfeqoGSJ9d_JFNSw-gpOeuU-ww2JKprk0vSvffk1GD7LplaN5C33Utiep-cOxuJp370OlbQQvkfao9tGOUC7a_mLptNr2YXN-ndw_XtxdldqnNZDmkhDM8A2rLQpmmqEpvcNFwjp8a0reDxUR1_01ARKo6yKRXnYBotFLYV5LvsZFm78P3bSGGo5zZock511I-hxrzIuRBQVRGFJap9H4InUy-8nSv_WSPU30LrKLT-FlqvhMbI0ap9bObU_gV-DUbgeAWooJUzXnXahn9OFFKKkkfucMlZIvo7lzweEfIv0r6HRA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>1353477088</pqid></control><display><type>article</type><title>Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation</title><source>IEEE Xplore (Online service)</source><creator>Ha Quang Minh ; Wiskott, Laurenz</creator><creatorcontrib>Ha Quang Minh ; Wiskott, Laurenz</creatorcontrib><description>We generalize the method of Slow Feature Analysis (SFA) for vector-valued functions of several variables and apply it to the problem of blind source separation, in particular to image separation. It is generally necessary to use multivariate SFA instead of univariate SFA for separating multi-dimensional signals. For the linear case, an exact mathematical analysis is given, which shows in particular that the sources are perfectly separated by SFA if and only if they and their first-order derivatives are uncorrelated. When the sources are correlated, we apply the following technique called Decorrelation Filtering: use a linear filter to decorrelate the sources and their derivatives in the given mixture, then apply the unmixing matrix obtained on the filtered mixtures to the original mixtures. If the filtered sources are perfectly separated by this matrix, so are the original sources. A decorrelation filter can be numerically obtained by solving a nonlinear optimization problem. This technique can also be applied to other linear separation methods, whose output signals are decorrelated, such as ICA. When there are more mixtures than sources, one can determine the actual number of sources by using a regularized version of SFA with decorrelation filtering. Extensive numerical experiments using SFA and ICA with decorrelation filtering, supported by mathematical analysis, demonstrate the potential of our methods for solving problems involving blind source separation.</description><identifier>ISSN: 1057-7149</identifier><identifier>EISSN: 1941-0042</identifier><identifier>DOI: 10.1109/TIP.2013.2257808</identifier><identifier>PMID: 23591489</identifier><identifier>CODEN: IIPRE4</identifier><language>eng</language><publisher>New York, NY: IEEE</publisher><subject>Applied sciences ; Blind source separation ; Detection, estimation, filtering, equalization, prediction ; Exact sciences and technology ; filtering ; generalized eigenvalue problem ; Image processing ; image separation ; independent component analysis ; Information, signal and communications theory ; Signal and communications theory ; Signal processing ; Signal, noise ; slow feature analysis ; Telecommunications and information theory</subject><ispartof>IEEE transactions on image processing, 2013-07, Vol.22 (7), p.2737-2750</ispartof><rights>2014 INIST-CNRS</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803</citedby><cites>FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/6497610$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>314,780,784,27924,27925,54796</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&amp;idt=27599764$$DView record in Pascal Francis$$Hfree_for_read</backlink><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/23591489$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Ha Quang Minh</creatorcontrib><creatorcontrib>Wiskott, Laurenz</creatorcontrib><title>Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation</title><title>IEEE transactions on image processing</title><addtitle>TIP</addtitle><addtitle>IEEE Trans Image Process</addtitle><description>We generalize the method of Slow Feature Analysis (SFA) for vector-valued functions of several variables and apply it to the problem of blind source separation, in particular to image separation. It is generally necessary to use multivariate SFA instead of univariate SFA for separating multi-dimensional signals. For the linear case, an exact mathematical analysis is given, which shows in particular that the sources are perfectly separated by SFA if and only if they and their first-order derivatives are uncorrelated. When the sources are correlated, we apply the following technique called Decorrelation Filtering: use a linear filter to decorrelate the sources and their derivatives in the given mixture, then apply the unmixing matrix obtained on the filtered mixtures to the original mixtures. If the filtered sources are perfectly separated by this matrix, so are the original sources. A decorrelation filter can be numerically obtained by solving a nonlinear optimization problem. This technique can also be applied to other linear separation methods, whose output signals are decorrelated, such as ICA. When there are more mixtures than sources, one can determine the actual number of sources by using a regularized version of SFA with decorrelation filtering. Extensive numerical experiments using SFA and ICA with decorrelation filtering, supported by mathematical analysis, demonstrate the potential of our methods for solving problems involving blind source separation.</description><subject>Applied sciences</subject><subject>Blind source separation</subject><subject>Detection, estimation, filtering, equalization, prediction</subject><subject>Exact sciences and technology</subject><subject>filtering</subject><subject>generalized eigenvalue problem</subject><subject>Image processing</subject><subject>image separation</subject><subject>independent component analysis</subject><subject>Information, signal and communications theory</subject><subject>Signal and communications theory</subject><subject>Signal processing</subject><subject>Signal, noise</subject><subject>slow feature analysis</subject><subject>Telecommunications and information theory</subject><issn>1057-7149</issn><issn>1941-0042</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2013</creationdate><recordtype>article</recordtype><recordid>eNpFkE1LxDAQhoMoft8FQXoRvHSdadOmOfq1KigK7slLSdOJRLLtmrSK_97ornqagXneF-Zh7ABhggjydHb7OMkA80mWFaKCao1to-SYAvBsPe5QiFQgl1tsJ4RXAOQFlptsK8sLibyS2-z5fnSDfVfeqoGSJ9d_JFNSw-gpOeuU-ww2JKprk0vSvffk1GD7LplaN5C33Utiep-cOxuJp370OlbQQvkfao9tGOUC7a_mLptNr2YXN-ndw_XtxdldqnNZDmkhDM8A2rLQpmmqEpvcNFwjp8a0reDxUR1_01ARKo6yKRXnYBotFLYV5LvsZFm78P3bSGGo5zZock511I-hxrzIuRBQVRGFJap9H4InUy-8nSv_WSPU30LrKLT-FlqvhMbI0ap9bObU_gV-DUbgeAWooJUzXnXahn9OFFKKkkfucMlZIvo7lzweEfIv0r6HRA</recordid><startdate>20130701</startdate><enddate>20130701</enddate><creator>Ha Quang Minh</creator><creator>Wiskott, Laurenz</creator><general>IEEE</general><general>Institute of Electrical and Electronics Engineers</general><scope>97E</scope><scope>RIA</scope><scope>RIE</scope><scope>IQODW</scope><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope></search><sort><creationdate>20130701</creationdate><title>Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation</title><author>Ha Quang Minh ; Wiskott, Laurenz</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Applied sciences</topic><topic>Blind source separation</topic><topic>Detection, estimation, filtering, equalization, prediction</topic><topic>Exact sciences and technology</topic><topic>filtering</topic><topic>generalized eigenvalue problem</topic><topic>Image processing</topic><topic>image separation</topic><topic>independent component analysis</topic><topic>Information, signal and communications theory</topic><topic>Signal and communications theory</topic><topic>Signal processing</topic><topic>Signal, noise</topic><topic>slow feature analysis</topic><topic>Telecommunications and information theory</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Ha Quang Minh</creatorcontrib><creatorcontrib>Wiskott, Laurenz</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005–Present</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998–Present</collection><collection>IEEE/IET Electronic Library (IEL)</collection><collection>Pascal-Francis</collection><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>IEEE transactions on image processing</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Ha Quang Minh</au><au>Wiskott, Laurenz</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation</atitle><jtitle>IEEE transactions on image processing</jtitle><stitle>TIP</stitle><addtitle>IEEE Trans Image Process</addtitle><date>2013-07-01</date><risdate>2013</risdate><volume>22</volume><issue>7</issue><spage>2737</spage><epage>2750</epage><pages>2737-2750</pages><issn>1057-7149</issn><eissn>1941-0042</eissn><coden>IIPRE4</coden><abstract>We generalize the method of Slow Feature Analysis (SFA) for vector-valued functions of several variables and apply it to the problem of blind source separation, in particular to image separation. It is generally necessary to use multivariate SFA instead of univariate SFA for separating multi-dimensional signals. For the linear case, an exact mathematical analysis is given, which shows in particular that the sources are perfectly separated by SFA if and only if they and their first-order derivatives are uncorrelated. When the sources are correlated, we apply the following technique called Decorrelation Filtering: use a linear filter to decorrelate the sources and their derivatives in the given mixture, then apply the unmixing matrix obtained on the filtered mixtures to the original mixtures. If the filtered sources are perfectly separated by this matrix, so are the original sources. A decorrelation filter can be numerically obtained by solving a nonlinear optimization problem. This technique can also be applied to other linear separation methods, whose output signals are decorrelated, such as ICA. When there are more mixtures than sources, one can determine the actual number of sources by using a regularized version of SFA with decorrelation filtering. Extensive numerical experiments using SFA and ICA with decorrelation filtering, supported by mathematical analysis, demonstrate the potential of our methods for solving problems involving blind source separation.</abstract><cop>New York, NY</cop><pub>IEEE</pub><pmid>23591489</pmid><doi>10.1109/TIP.2013.2257808</doi><tpages>14</tpages></addata></record>
fulltext fulltext
identifier ISSN: 1057-7149
ispartof IEEE transactions on image processing, 2013-07, Vol.22 (7), p.2737-2750
issn 1057-7149
1941-0042
language eng
recordid cdi_pascalfrancis_primary_27599764
source IEEE Xplore (Online service)
subjects Applied sciences
Blind source separation
Detection, estimation, filtering, equalization, prediction
Exact sciences and technology
filtering
generalized eigenvalue problem
Image processing
image separation
independent component analysis
Information, signal and communications theory
Signal and communications theory
Signal processing
Signal, noise
slow feature analysis
Telecommunications and information theory
title Multivariate Slow Feature Analysis and Decorrelation Filtering for Blind Source Separation
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T07%3A13%3A30IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_pasca&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Multivariate%20Slow%20Feature%20Analysis%20and%20Decorrelation%20Filtering%20for%20Blind%20Source%20Separation&rft.jtitle=IEEE%20transactions%20on%20image%20processing&rft.au=Ha%20Quang%20Minh&rft.date=2013-07-01&rft.volume=22&rft.issue=7&rft.spage=2737&rft.epage=2750&rft.pages=2737-2750&rft.issn=1057-7149&rft.eissn=1941-0042&rft.coden=IIPRE4&rft_id=info:doi/10.1109/TIP.2013.2257808&rft_dat=%3Cproquest_pasca%3E1353477088%3C/proquest_pasca%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-c396t-57f4200d65cfbb861b3fb4c14ebfdd74110c780c08e1a419b6a440fbc7a1d803%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1353477088&rft_id=info:pmid/23591489&rft_ieee_id=6497610&rfr_iscdi=true