The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Um esquema de reconhecimento de segmentação baseado em léxico no reconhecimento de nomes de cidades manuscritas em Bangla é proposto para a automação postal indiana. No esquema proposto, inicialmente é feita a binarização do documento de entrada e depois, para cuidar da caligrafia inclinada de diferentes indivíduos, é realizada uma técnica de correção de inclinação. Em seguida, devido às características da escrita de Bangla, um conceito de reservatório de água é aplicado para pré-segmentar os nomes das cidades com inclinação corrigida em possíveis componentes primitivos (personagens ou suas partes). Componentes pré-segmentados de um nome de cidade são então mesclados em caracteres possíveis para obter o melhor nome de cidade usando as informações do léxico. A fim de fundir esses componentes primitivos em caracteres e encontrar a segmentação ideal de caracteres, a programação dinâmica (DP) é aplicada usando a verossimilhança total dos caracteres de um nome de cidade como uma função objetivo. Para calcular a probabilidade de um caractere, a Função Discriminante Quadrática Modificada (MQDF) é usada. Os recursos utilizados no MQDF baseiam-se principalmente nos recursos direcionais dos pontos de contorno dos componentes. Testamos nosso sistema em 84 nomes de cidades bangla diferentes e 94.08% de precisão foi obtida no sistema proposto.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Umapada PAL, Kaushik ROY, Fumitaka KIMURA, "A Lexicon-Driven Handwritten City-Name Recognition Scheme for Indian Postal Automation" in IEICE TRANSACTIONS on Information,
vol. E92-D, no. 5, pp. 1146-1158, May 2009, doi: 10.1587/transinf.E92.D.1146.
Abstract: A lexicon-driven segmentation-recognition scheme on Bangla handwritten city-name recognition is proposed for Indian postal automation. In the proposed scheme, at first, binarization of the input document is done and then to take care of slanted handwriting of different individuals a slant correction technique is performed. Next, due to the script characteristics of Bangla, a water reservoir concept is applied to pre-segment the slant corrected city-names into possible primitive components (characters or its parts). Pre-segmented components of a city-name are then merged into possible characters to get the best city-name using the lexicon information. In order to merge these primitive components into characters and to find optimum character segmentation, dynamic programming (DP) is applied using total likelihood of the characters of a city-name as an objective function. To compute the likelihood of a character, Modified Quadratic Discriminant Function (MQDF) is used. The features used in the MQDF are mainly based on the directional features of the contour points of the components. We tested our system on 84 different Bangla city-names and 94.08% accuracy was obtained from the proposed system.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E92.D.1146/_p
Copiar
@ARTICLE{e92-d_5_1146,
author={Umapada PAL, Kaushik ROY, Fumitaka KIMURA, },
journal={IEICE TRANSACTIONS on Information},
title={A Lexicon-Driven Handwritten City-Name Recognition Scheme for Indian Postal Automation},
year={2009},
volume={E92-D},
number={5},
pages={1146-1158},
abstract={A lexicon-driven segmentation-recognition scheme on Bangla handwritten city-name recognition is proposed for Indian postal automation. In the proposed scheme, at first, binarization of the input document is done and then to take care of slanted handwriting of different individuals a slant correction technique is performed. Next, due to the script characteristics of Bangla, a water reservoir concept is applied to pre-segment the slant corrected city-names into possible primitive components (characters or its parts). Pre-segmented components of a city-name are then merged into possible characters to get the best city-name using the lexicon information. In order to merge these primitive components into characters and to find optimum character segmentation, dynamic programming (DP) is applied using total likelihood of the characters of a city-name as an objective function. To compute the likelihood of a character, Modified Quadratic Discriminant Function (MQDF) is used. The features used in the MQDF are mainly based on the directional features of the contour points of the components. We tested our system on 84 different Bangla city-names and 94.08% accuracy was obtained from the proposed system.},
keywords={},
doi={10.1587/transinf.E92.D.1146},
ISSN={1745-1361},
month={May},}
Copiar
TY - JOUR
TI - A Lexicon-Driven Handwritten City-Name Recognition Scheme for Indian Postal Automation
T2 - IEICE TRANSACTIONS on Information
SP - 1146
EP - 1158
AU - Umapada PAL
AU - Kaushik ROY
AU - Fumitaka KIMURA
PY - 2009
DO - 10.1587/transinf.E92.D.1146
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E92-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2009
AB - A lexicon-driven segmentation-recognition scheme on Bangla handwritten city-name recognition is proposed for Indian postal automation. In the proposed scheme, at first, binarization of the input document is done and then to take care of slanted handwriting of different individuals a slant correction technique is performed. Next, due to the script characteristics of Bangla, a water reservoir concept is applied to pre-segment the slant corrected city-names into possible primitive components (characters or its parts). Pre-segmented components of a city-name are then merged into possible characters to get the best city-name using the lexicon information. In order to merge these primitive components into characters and to find optimum character segmentation, dynamic programming (DP) is applied using total likelihood of the characters of a city-name as an objective function. To compute the likelihood of a character, Modified Quadratic Discriminant Function (MQDF) is used. The features used in the MQDF are mainly based on the directional features of the contour points of the components. We tested our system on 84 different Bangla city-names and 94.08% accuracy was obtained from the proposed system.
ER -