Feature identification for hybrid structural/statistical pattern classification
Identifieur interne : 000114 ( Istex/Curation ); précédent : 000113; suivant : 000115Feature identification for hybrid structural/statistical pattern classification
Auteurs : Henry S. Baird [États-Unis]Source :
- Computer Vision, Graphics and Image Processing [ 0734-189X ] ; 1987.
Abstract
A general technique for combining the strengths of structural shape analysis with statistical classification is proposed. The approach is to construct a function, called a feature identification mapping, from the representation generated by structural analysis to the one required for statistical classification. It is shown that if a certain continuity property holds for the parameterizations of the structural shape types, then it is possible to infer the mapping automatically. Inference is slow and heuristic, but is highly automated, controlled by only a few statistical parameters, and is applicable uniformly to all shape types. In addition, if the shape types are sufficiently elementary, the resulting mapping can be computed quickly using kD-trees. Large-scale statistically-significant trials, in the context of a mixed-font, variable-size optical character recognition (OCR) system, have shown that the technique is superior to simpler, fixed mappings, and is effective in generalizing common characteristics in mixtures of fonts.
Url:
DOI: 10.1016/S0734-189X(88)80042-2
Links toward previous steps (curation, corpus...)
- to stream Istex, to step Corpus: Pour aller vers cette notice dans l'étape Curation :000116
Links to Exploration step
ISTEX:EA084325D96A19F0E7289AAF7132A52FDDBDE6F5Le document en format XML
<record><TEI wicri:istexFullTextTei="biblStruct"><teiHeader><fileDesc><titleStmt><title xml:lang="en">Feature identification for hybrid structural/statistical pattern classification</title><author><name sortKey="Baird, Henry S" sort="Baird, Henry S" uniqKey="Baird H" first="Henry S." last="Baird">Henry S. Baird</name><affiliation wicri:level="1"><mods:affiliation>AT & T Bell Laboratories, Murray Hill, New Jersey 07974, USA</mods:affiliation><country xml:lang="fr">États-Unis</country><wicri:regionArea>AT & T Bell Laboratories, Murray Hill, New Jersey 07974</wicri:regionArea></affiliation></author></titleStmt><publicationStmt><idno type="wicri:source">ISTEX</idno><idno type="RBID">ISTEX:EA084325D96A19F0E7289AAF7132A52FDDBDE6F5</idno><date when="1988" year="1988">1988</date><idno type="doi">10.1016/S0734-189X(88)80042-2</idno><idno type="url">https://api.istex.fr/document/EA084325D96A19F0E7289AAF7132A52FDDBDE6F5/fulltext/pdf</idno><idno type="wicri:Area/Istex/Corpus">000116</idno><idno type="wicri:Area/Istex/Curation">000114</idno></publicationStmt><sourceDesc><biblStruct><analytic><title level="a" type="main" xml:lang="en">Feature identification for hybrid structural/statistical pattern classification</title><author><name sortKey="Baird, Henry S" sort="Baird, Henry S" uniqKey="Baird H" first="Henry S." last="Baird">Henry S. Baird</name><affiliation wicri:level="1"><mods:affiliation>AT & T Bell Laboratories, Murray Hill, New Jersey 07974, USA</mods:affiliation><country xml:lang="fr">États-Unis</country><wicri:regionArea>AT & T Bell Laboratories, Murray Hill, New Jersey 07974</wicri:regionArea></affiliation></author></analytic><monogr></monogr><series><title level="j">Computer Vision, Graphics and Image Processing</title><title level="j" type="abbrev">YCVIGP</title><idno type="ISSN">0734-189X</idno><imprint><publisher>ELSEVIER</publisher><date type="published" when="1987">1987</date><biblScope unit="volume">42</biblScope><biblScope unit="issue">3</biblScope><biblScope unit="page" from="318">318</biblScope><biblScope unit="page" to="333">333</biblScope></imprint><idno type="ISSN">0734-189X</idno></series><idno type="istex">EA084325D96A19F0E7289AAF7132A52FDDBDE6F5</idno><idno type="DOI">10.1016/S0734-189X(88)80042-2</idno><idno type="PII">S0734-189X(88)80042-2</idno><idno type="ArticleID">88800422</idno></biblStruct></sourceDesc><seriesStmt><idno type="ISSN">0734-189X</idno></seriesStmt></fileDesc><profileDesc><textClass></textClass><langUsage><language ident="en">en</language></langUsage></profileDesc></teiHeader><front><div type="abstract" xml:lang="en">A general technique for combining the strengths of structural shape analysis with statistical classification is proposed. The approach is to construct a function, called a feature identification mapping, from the representation generated by structural analysis to the one required for statistical classification. It is shown that if a certain continuity property holds for the parameterizations of the structural shape types, then it is possible to infer the mapping automatically. Inference is slow and heuristic, but is highly automated, controlled by only a few statistical parameters, and is applicable uniformly to all shape types. In addition, if the shape types are sufficiently elementary, the resulting mapping can be computed quickly using kD-trees. Large-scale statistically-significant trials, in the context of a mixed-font, variable-size optical character recognition (OCR) system, have shown that the technique is superior to simpler, fixed mappings, and is effective in generalizing common characteristics in mixtures of fonts.</div></front></TEI></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/Istex/Curation
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000114 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Istex/Curation/biblio.hfd -nk 000114 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Ticri/CIDE
|area= OcrV1
|flux= Istex
|étape= Curation
|type= RBID
|clé= ISTEX:EA084325D96A19F0E7289AAF7132A52FDDBDE6F5
|texte= Feature identification for hybrid structural/statistical pattern classification
}}
| This area was generated with Dilib version V0.6.32. |  |