Hand-Written Text Recognition Based on a New Formulation
Identifieur interne : 00DC72 ( Main/Merge ); précédent : 00DC71; suivant : 00DC73Hand-Written Text Recognition Based on a New Formulation
Auteurs : Y. Gong ; A. BoyerSource :
Abstract
In this paper, hand-written word recognition problem is formulated as two steps. In the first step, the plausibility of observing each character is computed as function of sample index of a line. In the second step the word recognition is achieved by finding the word (sequence of characters) which maximizes the sum of plausibilities of individual characters which make up the word. For the second step we propose an efficient algorithm which makes use of peaks of plausibility functions and solve the maximization process by two embedded search processes, using dynamic programmings : finding the best path connecting peaks in the plausibility functions of two successive characters, and finding the best transition sample index for two given peaks. In a preliminary experimentation, using template matching for character plausibility estimation we obtained 96========percnt; recognition rate on a hand-written-like font.
Links toward previous steps (curation, corpus...)
- to stream Crin, to step Corpus: 000D80
- to stream Crin, to step Curation: 000D80
- to stream Crin, to step Checkpoint: 003652
Links to Exploration step
CRIN:gong92aLe document en format XML
<record><TEI><teiHeader><fileDesc><titleStmt><title xml:lang="en" wicri:score="460">Hand-Written Text Recognition Based on a New Formulation</title></titleStmt><publicationStmt><idno type="RBID">CRIN:gong92a</idno><date when="1992" year="1992">1992</date><idno type="wicri:Area/Crin/Corpus">000D80</idno><idno type="wicri:Area/Crin/Curation">000D80</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Curation">000D80</idno><idno type="wicri:Area/Crin/Checkpoint">003652</idno><idno type="wicri:explorRef" wicri:stream="Crin" wicri:step="Checkpoint">003652</idno><idno type="wicri:Area/Main/Merge">00DC72</idno></publicationStmt><sourceDesc><biblStruct><analytic><title xml:lang="en">Hand-Written Text Recognition Based on a New Formulation</title><author><name sortKey="Gong, Y" sort="Gong, Y" uniqKey="Gong Y" first="Y." last="Gong">Y. Gong</name></author><author><name sortKey="Boyer, A" sort="Boyer, A" uniqKey="Boyer A" first="A." last="Boyer">A. Boyer</name></author></analytic></biblStruct></sourceDesc></fileDesc><profileDesc><textClass></textClass></profileDesc></teiHeader><front><div type="abstract" xml:lang="en" wicri:score="3167">In this paper, hand-written word recognition problem is formulated as two steps. In the first step, the plausibility of observing each character is computed as function of sample index of a line. In the second step the word recognition is achieved by finding the word (sequence of characters) which maximizes the sum of plausibilities of individual characters which make up the word. For the second step we propose an efficient algorithm which makes use of peaks of plausibility functions and solve the maximization process by two embedded search processes, using dynamic programmings : finding the best path connecting peaks in the plausibility functions of two successive characters, and finding the best transition sample index for two given peaks. In a preliminary experimentation, using template matching for character plausibility estimation we obtained 96========percnt; recognition rate on a hand-written-like font.</div></front></TEI></record>Pour manipuler ce document sous Unix (Dilib)
EXPLOR_STEP=$WICRI_ROOT/Wicri/Lorraine/explor/InforLorV4/Data/Main/Merge
HfdSelect -h $EXPLOR_STEP/biblio.hfd -nk 00DC72 | SxmlIndent | more
Ou
HfdSelect -h $EXPLOR_AREA/Data/Main/Merge/biblio.hfd -nk 00DC72 | SxmlIndent | more
Pour mettre un lien sur cette page dans le réseau Wicri
{{Explor lien
|wiki= Wicri/Lorraine
|area= InforLorV4
|flux= Main
|étape= Merge
|type= RBID
|clé= CRIN:gong92a
|texte= Hand-Written Text Recognition Based on a New Formulation
}}
| This area was generated with Dilib version V0.6.33. |  |