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
A detecção de texto é uma etapa crucial de pré-processamento no reconhecimento óptico de caracteres (OCR) para o reconhecimento preciso de texto, incluindo fontes e caracteres manuscritos, em documentos. Embora as atuais ferramentas de detecção de texto baseadas em aprendizagem profunda possam detectar regiões de texto com alta precisão, elas geralmente tratam múltiplas linhas de texto como uma única região. Para realizar o reconhecimento de caracteres baseado em linhas, é necessário dividir o texto em linhas individuais, o que requer uma técnica de detecção de linhas. Este artigo se concentra no desenvolvimento de uma nova abordagem para detecção de linha única em OCR que é baseada no modelo existente de reconhecimento de região de caracteres para detecção de texto (CRAFT) e incorpora uma rede neural profunda especializada em segmentação de linha. No entanto, este novo método ainda pode detectar múltiplas linhas como uma única região de texto quando estiver presente texto de múltiplas linhas com espaçamento estreito. Para resolver isso, também introduzimos um algoritmo de pós-processamento para detectar regiões de texto único usando a saída da segmentação de linha única. Nosso método proposto detecta com sucesso linhas únicas, mesmo em texto multilinha com espaçamento estreito entre linhas e, portanto, melhora a precisão do OCR.
Chee Siang LEOW
University of Yamanashi
Hideaki YAJIMA
University of Yamanashi
Tomoki KITAGAWA
University of Yamanashi
Hiromitsu NISHIZAKI
University of Yamanashi
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
Chee Siang LEOW, Hideaki YAJIMA, Tomoki KITAGAWA, Hiromitsu NISHIZAKI, "Single-Line Text Detection in Multi-Line Text with Narrow Spacing for Line-Based Character Recognition" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 12, pp. 2097-2106, December 2023, doi: 10.1587/transinf.2023EDP7070.
Abstract: Text detection is a crucial pre-processing step in optical character recognition (OCR) for the accurate recognition of text, including both fonts and handwritten characters, in documents. While current deep learning-based text detection tools can detect text regions with high accuracy, they often treat multiple lines of text as a single region. To perform line-based character recognition, it is necessary to divide the text into individual lines, which requires a line detection technique. This paper focuses on the development of a new approach to single-line detection in OCR that is based on the existing Character Region Awareness For Text detection (CRAFT) model and incorporates a deep neural network specialized in line segmentation. However, this new method may still detect multiple lines as a single text region when multi-line text with narrow spacing is present. To address this, we also introduce a post-processing algorithm to detect single text regions using the output of the single-line segmentation. Our proposed method successfully detects single lines, even in multi-line text with narrow line spacing, and hence improves the accuracy of OCR.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2023EDP7070/_p
Copiar
@ARTICLE{e106-d_12_2097,
author={Chee Siang LEOW, Hideaki YAJIMA, Tomoki KITAGAWA, Hiromitsu NISHIZAKI, },
journal={IEICE TRANSACTIONS on Information},
title={Single-Line Text Detection in Multi-Line Text with Narrow Spacing for Line-Based Character Recognition},
year={2023},
volume={E106-D},
number={12},
pages={2097-2106},
abstract={Text detection is a crucial pre-processing step in optical character recognition (OCR) for the accurate recognition of text, including both fonts and handwritten characters, in documents. While current deep learning-based text detection tools can detect text regions with high accuracy, they often treat multiple lines of text as a single region. To perform line-based character recognition, it is necessary to divide the text into individual lines, which requires a line detection technique. This paper focuses on the development of a new approach to single-line detection in OCR that is based on the existing Character Region Awareness For Text detection (CRAFT) model and incorporates a deep neural network specialized in line segmentation. However, this new method may still detect multiple lines as a single text region when multi-line text with narrow spacing is present. To address this, we also introduce a post-processing algorithm to detect single text regions using the output of the single-line segmentation. Our proposed method successfully detects single lines, even in multi-line text with narrow line spacing, and hence improves the accuracy of OCR.},
keywords={},
doi={10.1587/transinf.2023EDP7070},
ISSN={1745-1361},
month={December},}
Copiar
TY - JOUR
TI - Single-Line Text Detection in Multi-Line Text with Narrow Spacing for Line-Based Character Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 2097
EP - 2106
AU - Chee Siang LEOW
AU - Hideaki YAJIMA
AU - Tomoki KITAGAWA
AU - Hiromitsu NISHIZAKI
PY - 2023
DO - 10.1587/transinf.2023EDP7070
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2023
AB - Text detection is a crucial pre-processing step in optical character recognition (OCR) for the accurate recognition of text, including both fonts and handwritten characters, in documents. While current deep learning-based text detection tools can detect text regions with high accuracy, they often treat multiple lines of text as a single region. To perform line-based character recognition, it is necessary to divide the text into individual lines, which requires a line detection technique. This paper focuses on the development of a new approach to single-line detection in OCR that is based on the existing Character Region Awareness For Text detection (CRAFT) model and incorporates a deep neural network specialized in line segmentation. However, this new method may still detect multiple lines as a single text region when multi-line text with narrow spacing is present. To address this, we also introduce a post-processing algorithm to detect single text regions using the output of the single-line segmentation. Our proposed method successfully detects single lines, even in multi-line text with narrow line spacing, and hence improves the accuracy of OCR.
ER -