Novel histograms kernels with structural properties
•We study the space where histograms lie.•We introduce some intuitive and desirable structural properties for measures.•A new similarity measure for comparing histograms is proposed.•We show that the proposed similarity is a conditionally positive definite kernel.•Experiments on face recognition and...
Gespeichert in:
| Veröffentlicht in: | Pattern recognition letters 2015-12, Vol.68, p.146-152 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 152 |
|---|---|
| container_issue | |
| container_start_page | 146 |
| container_title | Pattern recognition letters |
| container_volume | 68 |
| creator | Correa-Morris, Jyrko Martínez-Díaz, Yoanna Hernández, Noslen Méndez-Vázquez, Heydi |
| description | •We study the space where histograms lie.•We introduce some intuitive and desirable structural properties for measures.•A new similarity measure for comparing histograms is proposed.•We show that the proposed similarity is a conditionally positive definite kernel.•Experiments on face recognition and image retrieval were done.
This paper introduces a new similarity measure for comparing histograms, named Weighted Distribution Matching, which bases the comparison not only in the specific bin values but also in the shape of the histograms. It is proved that the proposed similarity is a conditionally positive definite kernel. The space where histograms lie is studied, and some intuitively desirable structural properties are introduced. The most representative measures of the state of art were compared on the basis of these properties. Experiments conducted on face recognition and image retrieval validate the proposal. |
| doi_str_mv | 10.1016/j.patrec.2015.09.005 |
| format | Article |
| fullrecord | <record><control><sourceid>elsevier_hal_p</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_04804108v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><els_id>S0167865515003025</els_id><sourcerecordid>S0167865515003025</sourcerecordid><originalsourceid>FETCH-LOGICAL-c289t-17545f99962baf073fd286433aff9d9ae5e583ff876ad27d17b34db0e95533a73</originalsourceid><addsrcrecordid>eNp9kEFLwzAYhoMoOKf_wEOvHlq_NEmTXIQxdBOGXvQcsvaLy-zWkmQT_70dFY-ePnh53he-h5BbCgUFWt1vi96mgHVRAhUF6AJAnJEJVbLMJeP8nEwGTOaqEuKSXMW4BYCKaTUh7KU7YpttfEzdR7C7mH1i2GMbsy-fNllM4VCnQ7Bt1oeux5A8xmty4Wwb8eb3Tsn70-PbfJmvXhfP89kqr0ulU06l4MJpratybR1I5ppSVZwx65xutEWBQjHnlKxsU8qGyjXjzRpQCzFAkk3J3bi7sa3pg9_Z8G06681ytjKnDLgCTkEd6cDyka1DF2NA91egYE6SzNaMksxJkgFtBklD7WGsDR_j0WMwsfa4r7HxA5pM0_n_B34AY2pxvw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Novel histograms kernels with structural properties</title><source>Elsevier ScienceDirect Journals</source><creator>Correa-Morris, Jyrko ; Martínez-Díaz, Yoanna ; Hernández, Noslen ; Méndez-Vázquez, Heydi</creator><creatorcontrib>Correa-Morris, Jyrko ; Martínez-Díaz, Yoanna ; Hernández, Noslen ; Méndez-Vázquez, Heydi</creatorcontrib><description>•We study the space where histograms lie.•We introduce some intuitive and desirable structural properties for measures.•A new similarity measure for comparing histograms is proposed.•We show that the proposed similarity is a conditionally positive definite kernel.•Experiments on face recognition and image retrieval were done.
This paper introduces a new similarity measure for comparing histograms, named Weighted Distribution Matching, which bases the comparison not only in the specific bin values but also in the shape of the histograms. It is proved that the proposed similarity is a conditionally positive definite kernel. The space where histograms lie is studied, and some intuitively desirable structural properties are introduced. The most representative measures of the state of art were compared on the basis of these properties. Experiments conducted on face recognition and image retrieval validate the proposal.</description><identifier>ISSN: 0167-8655</identifier><identifier>EISSN: 1872-7344</identifier><identifier>DOI: 10.1016/j.patrec.2015.09.005</identifier><language>eng</language><publisher>Elsevier B.V</publisher><subject>Face recognition ; Histogram similarity ; Image retrieval ; Kernel function ; Machine Learning ; Statistics</subject><ispartof>Pattern recognition letters, 2015-12, Vol.68, p.146-152</ispartof><rights>2015 Elsevier B.V.</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><cites>FETCH-LOGICAL-c289t-17545f99962baf073fd286433aff9d9ae5e583ff876ad27d17b34db0e95533a73</cites></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://dx.doi.org/10.1016/j.patrec.2015.09.005$$EHTML$$P50$$Gelsevier$$H</linktohtml><link.rule.ids>230,314,777,781,882,3537,27905,27906,45976</link.rule.ids><backlink>$$Uhttps://hal.science/hal-04804108$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Correa-Morris, Jyrko</creatorcontrib><creatorcontrib>Martínez-Díaz, Yoanna</creatorcontrib><creatorcontrib>Hernández, Noslen</creatorcontrib><creatorcontrib>Méndez-Vázquez, Heydi</creatorcontrib><title>Novel histograms kernels with structural properties</title><title>Pattern recognition letters</title><description>•We study the space where histograms lie.•We introduce some intuitive and desirable structural properties for measures.•A new similarity measure for comparing histograms is proposed.•We show that the proposed similarity is a conditionally positive definite kernel.•Experiments on face recognition and image retrieval were done.
This paper introduces a new similarity measure for comparing histograms, named Weighted Distribution Matching, which bases the comparison not only in the specific bin values but also in the shape of the histograms. It is proved that the proposed similarity is a conditionally positive definite kernel. The space where histograms lie is studied, and some intuitively desirable structural properties are introduced. The most representative measures of the state of art were compared on the basis of these properties. Experiments conducted on face recognition and image retrieval validate the proposal.</description><subject>Face recognition</subject><subject>Histogram similarity</subject><subject>Image retrieval</subject><subject>Kernel function</subject><subject>Machine Learning</subject><subject>Statistics</subject><issn>0167-8655</issn><issn>1872-7344</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2015</creationdate><recordtype>article</recordtype><recordid>eNp9kEFLwzAYhoMoOKf_wEOvHlq_NEmTXIQxdBOGXvQcsvaLy-zWkmQT_70dFY-ePnh53he-h5BbCgUFWt1vi96mgHVRAhUF6AJAnJEJVbLMJeP8nEwGTOaqEuKSXMW4BYCKaTUh7KU7YpttfEzdR7C7mH1i2GMbsy-fNllM4VCnQ7Bt1oeux5A8xmty4Wwb8eb3Tsn70-PbfJmvXhfP89kqr0ulU06l4MJpratybR1I5ppSVZwx65xutEWBQjHnlKxsU8qGyjXjzRpQCzFAkk3J3bi7sa3pg9_Z8G06681ytjKnDLgCTkEd6cDyka1DF2NA91egYE6SzNaMksxJkgFtBklD7WGsDR_j0WMwsfa4r7HxA5pM0_n_B34AY2pxvw</recordid><startdate>20151215</startdate><enddate>20151215</enddate><creator>Correa-Morris, Jyrko</creator><creator>Martínez-Díaz, Yoanna</creator><creator>Hernández, Noslen</creator><creator>Méndez-Vázquez, Heydi</creator><general>Elsevier B.V</general><general>Elsevier</general><scope>AAYXX</scope><scope>CITATION</scope><scope>1XC</scope><scope>VOOES</scope></search><sort><creationdate>20151215</creationdate><title>Novel histograms kernels with structural properties</title><author>Correa-Morris, Jyrko ; Martínez-Díaz, Yoanna ; Hernández, Noslen ; Méndez-Vázquez, Heydi</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c289t-17545f99962baf073fd286433aff9d9ae5e583ff876ad27d17b34db0e95533a73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2015</creationdate><topic>Face recognition</topic><topic>Histogram similarity</topic><topic>Image retrieval</topic><topic>Kernel function</topic><topic>Machine Learning</topic><topic>Statistics</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Correa-Morris, Jyrko</creatorcontrib><creatorcontrib>Martínez-Díaz, Yoanna</creatorcontrib><creatorcontrib>Hernández, Noslen</creatorcontrib><creatorcontrib>Méndez-Vázquez, Heydi</creatorcontrib><collection>CrossRef</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>Pattern recognition letters</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Correa-Morris, Jyrko</au><au>Martínez-Díaz, Yoanna</au><au>Hernández, Noslen</au><au>Méndez-Vázquez, Heydi</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Novel histograms kernels with structural properties</atitle><jtitle>Pattern recognition letters</jtitle><date>2015-12-15</date><risdate>2015</risdate><volume>68</volume><spage>146</spage><epage>152</epage><pages>146-152</pages><issn>0167-8655</issn><eissn>1872-7344</eissn><abstract>•We study the space where histograms lie.•We introduce some intuitive and desirable structural properties for measures.•A new similarity measure for comparing histograms is proposed.•We show that the proposed similarity is a conditionally positive definite kernel.•Experiments on face recognition and image retrieval were done.
This paper introduces a new similarity measure for comparing histograms, named Weighted Distribution Matching, which bases the comparison not only in the specific bin values but also in the shape of the histograms. It is proved that the proposed similarity is a conditionally positive definite kernel. The space where histograms lie is studied, and some intuitively desirable structural properties are introduced. The most representative measures of the state of art were compared on the basis of these properties. Experiments conducted on face recognition and image retrieval validate the proposal.</abstract><pub>Elsevier B.V</pub><doi>10.1016/j.patrec.2015.09.005</doi><tpages>7</tpages><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 0167-8655 |
| ispartof | Pattern recognition letters, 2015-12, Vol.68, p.146-152 |
| issn | 0167-8655 1872-7344 |
| language | eng |
| recordid | cdi_hal_primary_oai_HAL_hal_04804108v1 |
| source | Elsevier ScienceDirect Journals |
| subjects | Face recognition Histogram similarity Image retrieval Kernel function Machine Learning Statistics |
| title | Novel histograms kernels with structural properties |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-18T21%3A08%3A11IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-elsevier_hal_p&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Novel%20histograms%20kernels%20with%20structural%20properties&rft.jtitle=Pattern%20recognition%20letters&rft.au=Correa-Morris,%20Jyrko&rft.date=2015-12-15&rft.volume=68&rft.spage=146&rft.epage=152&rft.pages=146-152&rft.issn=0167-8655&rft.eissn=1872-7344&rft_id=info:doi/10.1016/j.patrec.2015.09.005&rft_dat=%3Celsevier_hal_p%3ES0167865515003025%3C/elsevier_hal_p%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_els_id=S0167865515003025&rfr_iscdi=true |