High accuracy optical character recognition using neural networks with centroid dithering
Identifieur interne : 000A10 ( PascalFrancis/Checkpoint ); précédent : 000A09; suivant : 000A11High accuracy optical character recognition using neural networks with centroid dithering
Auteurs : H. I. Avi-Itzhak [États-Unis] ; T. A. Diep [États-Unis] ; H. GarlandSource :
- IEEE transactions on pattern analysis and machine intelligence [ 0162-8828 ] ; 1995.
Descripteurs français
- Pascal (Inist)
- Wicri :
- topic : Base de données.
English descriptors
- KwdEn :
Abstract
Optical character recognition (OCR) refers to a process whereby printed documents are transformed into ASCII Files for the purpose of compact storage, editing, fast retrieval, and other File manipulations through the use of a computer. The recognition stage of an OCR process is made difficult by added noise, image distortion, and the various character typefaces, sizes, and fonts that a document may have. In this study a neural network approach is introduced to perform high accuracy recognition on multi-size and multi-font characters; a novel centroid-dithering training process with a low noise-sensitivity normalization procedure is used to achieve high accuracy results. The study consists of two parts. The first part focuses on single size and single font characters, and a two-layered neural network is trained to recognize the full set of 94 ASCII character images in 12-pt Courier font. The second part trades accuracy for additional font and size capability, and a larger two-layered neural network is trained to recognize the full set of 94 ASCII character images for all point sizes from 8 to 32 and for 12 commonly used fonts. The performance of these two networks is evaluated based on a database of more than one million character images from the testing data set
Affiliations:
Links toward previous steps (curation, corpus...)
Links to Exploration step
Pascal:95-0146490Le document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" level="a">High accuracy optical character recognition using neural networks with centroid dithering</title>
<author><name sortKey="Avi Itzhak, H I" sort="Avi Itzhak, H I" uniqKey="Avi Itzhak H" first="H. I." last="Avi-Itzhak">H. I. Avi-Itzhak</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Stanford univ., dep. electrical eng.</s1>
<s2>Stanford CA 94305</s2>
<s3>USA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Stanford CA 94305</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Diep, T A" sort="Diep, T A" uniqKey="Diep T" first="T. A." last="Diep">T. A. Diep</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Stanford univ., dep. electrical eng.</s1>
<s2>Stanford CA 94305</s2>
<s3>USA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Stanford CA 94305</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Garland, H" sort="Garland, H" uniqKey="Garland H" first="H." last="Garland">H. Garland</name>
</author>
</titleStmt>
<publicationStmt><idno type="wicri:source">INIST</idno>
<idno type="inist">95-0146490</idno>
<date when="1995">1995</date>
<idno type="stanalyst">PASCAL 95-0146490 INIST</idno>
<idno type="RBID">Pascal:95-0146490</idno>
<idno type="wicri:Area/PascalFrancis/Corpus">000A73</idno>
<idno type="wicri:Area/PascalFrancis/Curation">000926</idno>
<idno type="wicri:Area/PascalFrancis/Checkpoint">000A10</idno>
</publicationStmt>
<sourceDesc><biblStruct><analytic><title xml:lang="en" level="a">High accuracy optical character recognition using neural networks with centroid dithering</title>
<author><name sortKey="Avi Itzhak, H I" sort="Avi Itzhak, H I" uniqKey="Avi Itzhak H" first="H. I." last="Avi-Itzhak">H. I. Avi-Itzhak</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Stanford univ., dep. electrical eng.</s1>
<s2>Stanford CA 94305</s2>
<s3>USA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Stanford CA 94305</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Diep, T A" sort="Diep, T A" uniqKey="Diep T" first="T. A." last="Diep">T. A. Diep</name>
<affiliation wicri:level="1"><inist:fA14 i1="01"><s1>Stanford univ., dep. electrical eng.</s1>
<s2>Stanford CA 94305</s2>
<s3>USA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</inist:fA14>
<country>États-Unis</country>
<wicri:noRegion>Stanford CA 94305</wicri:noRegion>
</affiliation>
</author>
<author><name sortKey="Garland, H" sort="Garland, H" uniqKey="Garland H" first="H." last="Garland">H. Garland</name>
</author>
</analytic>
<series><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
<imprint><date when="1995">1995</date>
</imprint>
</series>
</biblStruct>
</sourceDesc>
<seriesStmt><title level="j" type="main">IEEE transactions on pattern analysis and machine intelligence</title>
<title level="j" type="abbreviated">IEEE trans. pattern anal. mach. intell.</title>
<idno type="ISSN">0162-8828</idno>
</seriesStmt>
</fileDesc>
<profileDesc><textClass><keywords scheme="KwdEn" xml:lang="en"><term>Character recognition</term>
<term>Character set</term>
<term>Database</term>
<term>Neural network</term>
<term>Noisy image</term>
<term>Pattern recognition</term>
</keywords>
<keywords scheme="Pascal" xml:lang="fr"><term>Reconnaissance caractère</term>
<term>Reconnaissance forme</term>
<term>Image bruitée</term>
<term>Réseau neuronal</term>
<term>Base donnée</term>
<term>Jeu caractère</term>
<term>Optical character recognition</term>
<term>Multi font preprocessing</term>
<term>Multi size preprocessing</term>
</keywords>
<keywords scheme="Wicri" type="topic" xml:lang="fr"><term>Base de données</term>
</keywords>
</textClass>
</profileDesc>
</teiHeader>
<front><div type="abstract" xml:lang="en">Optical character recognition (OCR) refers to a process whereby printed documents are transformed into ASCII Files for the purpose of compact storage, editing, fast retrieval, and other File manipulations through the use of a computer. The recognition stage of an OCR process is made difficult by added noise, image distortion, and the various character typefaces, sizes, and fonts that a document may have. In this study a neural network approach is introduced to perform high accuracy recognition on multi-size and multi-font characters; a novel centroid-dithering training process with a low noise-sensitivity normalization procedure is used to achieve high accuracy results. The study consists of two parts. The first part focuses on single size and single font characters, and a two-layered neural network is trained to recognize the full set of 94 ASCII character images in 12-pt Courier font. The second part trades accuracy for additional font and size capability, and a larger two-layered neural network is trained to recognize the full set of 94 ASCII character images for all point sizes from 8 to 32 and for 12 commonly used fonts. The performance of these two networks is evaluated based on a database of more than one million character images from the testing data set</div>
</front>
</TEI>
<inist><standard h6="B"><pA><fA01 i1="01" i2="1"><s0>0162-8828</s0>
</fA01>
<fA02 i1="01"><s0>ITPIDJ</s0>
</fA02>
<fA03 i2="1"><s0>IEEE trans. pattern anal. mach. intell.</s0>
</fA03>
<fA05><s2>17</s2>
</fA05>
<fA06><s2>2</s2>
</fA06>
<fA08 i1="01" i2="1" l="ENG"><s1>High accuracy optical character recognition using neural networks with centroid dithering</s1>
</fA08>
<fA11 i1="01" i2="1"><s1>AVI-ITZHAK (H. I.)</s1>
</fA11>
<fA11 i1="02" i2="1"><s1>DIEP (T. A.)</s1>
</fA11>
<fA11 i1="03" i2="1"><s1>GARLAND (H.)</s1>
</fA11>
<fA14 i1="01"><s1>Stanford univ., dep. electrical eng.</s1>
<s2>Stanford CA 94305</s2>
<s3>USA</s3>
<sZ>1 aut.</sZ>
<sZ>2 aut.</sZ>
</fA14>
<fA20><s1>218-224</s1>
</fA20>
<fA21><s1>1995</s1>
</fA21>
<fA23 i1="01"><s0>ENG</s0>
</fA23>
<fA43 i1="01"><s1>INIST</s1>
<s2>222T</s2>
<s5>354000059459030130</s5>
</fA43>
<fA44><s0>0000</s0>
</fA44>
<fA45><s0>11 ref.</s0>
</fA45>
<fA47 i1="01" i2="1"><s0>95-0146490</s0>
</fA47>
<fA60><s1>P</s1>
<s3>CR</s3>
</fA60>
<fA61><s0>A</s0>
</fA61>
<fA64 i1="01" i2="1"><s0>IEEE transactions on pattern analysis and machine intelligence</s0>
</fA64>
<fA66 i1="01"><s0>USA</s0>
</fA66>
<fC01 i1="01" l="ENG"><s0>Optical character recognition (OCR) refers to a process whereby printed documents are transformed into ASCII Files for the purpose of compact storage, editing, fast retrieval, and other File manipulations through the use of a computer. The recognition stage of an OCR process is made difficult by added noise, image distortion, and the various character typefaces, sizes, and fonts that a document may have. In this study a neural network approach is introduced to perform high accuracy recognition on multi-size and multi-font characters; a novel centroid-dithering training process with a low noise-sensitivity normalization procedure is used to achieve high accuracy results. The study consists of two parts. The first part focuses on single size and single font characters, and a two-layered neural network is trained to recognize the full set of 94 ASCII character images in 12-pt Courier font. The second part trades accuracy for additional font and size capability, and a larger two-layered neural network is trained to recognize the full set of 94 ASCII character images for all point sizes from 8 to 32 and for 12 commonly used fonts. The performance of these two networks is evaluated based on a database of more than one million character images from the testing data set</s0>
</fC01>
<fC02 i1="01" i2="X"><s0>001D02C03</s0>
</fC02>
<fC02 i1="02" i2="X"><s0>001D02C06</s0>
</fC02>
<fC03 i1="01" i2="X" l="FRE"><s0>Reconnaissance caractère</s0>
<s5>67</s5>
</fC03>
<fC03 i1="01" i2="X" l="ENG"><s0>Character recognition</s0>
<s5>67</s5>
</fC03>
<fC03 i1="01" i2="X" l="SPA"><s0>Reconocimiento carácter</s0>
<s5>67</s5>
</fC03>
<fC03 i1="02" i2="X" l="FRE"><s0>Reconnaissance forme</s0>
<s5>68</s5>
</fC03>
<fC03 i1="02" i2="X" l="ENG"><s0>Pattern recognition</s0>
<s5>68</s5>
</fC03>
<fC03 i1="02" i2="X" l="GER"><s0>Mustererkennung</s0>
<s5>68</s5>
</fC03>
<fC03 i1="02" i2="X" l="SPA"><s0>Reconocimiento patrón</s0>
<s5>68</s5>
</fC03>
<fC03 i1="03" i2="X" l="FRE"><s0>Image bruitée</s0>
<s5>69</s5>
</fC03>
<fC03 i1="03" i2="X" l="ENG"><s0>Noisy image</s0>
<s5>69</s5>
</fC03>
<fC03 i1="03" i2="X" l="SPA"><s0>Imagen sonora</s0>
<s5>69</s5>
</fC03>
<fC03 i1="04" i2="X" l="FRE"><s0>Réseau neuronal</s0>
<s5>70</s5>
</fC03>
<fC03 i1="04" i2="X" l="ENG"><s0>Neural network</s0>
<s5>70</s5>
</fC03>
<fC03 i1="04" i2="X" l="SPA"><s0>Red neuronal</s0>
<s5>70</s5>
</fC03>
<fC03 i1="05" i2="X" l="FRE"><s0>Base donnée</s0>
<s5>71</s5>
</fC03>
<fC03 i1="05" i2="X" l="ENG"><s0>Database</s0>
<s5>71</s5>
</fC03>
<fC03 i1="05" i2="X" l="SPA"><s0>Base dato</s0>
<s5>71</s5>
</fC03>
<fC03 i1="06" i2="X" l="FRE"><s0>Jeu caractère</s0>
<s5>72</s5>
</fC03>
<fC03 i1="06" i2="X" l="ENG"><s0>Character set</s0>
<s5>72</s5>
</fC03>
<fC03 i1="06" i2="X" l="SPA"><s0>Juego caracter</s0>
<s5>72</s5>
</fC03>
<fC03 i1="07" i2="X" l="FRE"><s0>Optical character recognition</s0>
<s4>INC</s4>
<s5>90</s5>
</fC03>
<fC03 i1="08" i2="X" l="FRE"><s0>Multi font preprocessing</s0>
<s4>INC</s4>
<s5>91</s5>
</fC03>
<fC03 i1="09" i2="X" l="FRE"><s0>Multi size preprocessing</s0>
<s4>INC</s4>
<s5>92</s5>
</fC03>
<fN21><s1>088</s1>
</fN21>
</pA>
</standard>
</inist>
<affiliations><list><country><li>États-Unis</li>
</country>
</list>
<tree><noCountry><name sortKey="Garland, H" sort="Garland, H" uniqKey="Garland H" first="H." last="Garland">H. Garland</name>
</noCountry>
<country name="États-Unis"><noRegion><name sortKey="Avi Itzhak, H I" sort="Avi Itzhak, H I" uniqKey="Avi Itzhak H" first="H. I." last="Avi-Itzhak">H. I. Avi-Itzhak</name>
</noRegion>
<name sortKey="Diep, T A" sort="Diep, T A" uniqKey="Diep T" first="T. A." last="Diep">T. A. Diep</name>
</country>
</tree>
</affiliations>
</record>
Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Ticri/CIDE/explor/OcrV1/Data/PascalFrancis/Checkpoint
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 000A10 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/PascalFrancis/Checkpoint/biblio.hfd -nk 000A10 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien |wiki= Ticri/CIDE |area= OcrV1 |flux= PascalFrancis |étape= Checkpoint |type= RBID |clé= Pascal:95-0146490 |texte= High accuracy optical character recognition using neural networks with centroid dithering }}
This area was generated with Dilib version V0.6.32. |