Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation

Accurate layout analysis without subsequent text-line segmentation remains an ongoing challenge, especially when facing the Kangyur, a kind of historical Tibetan document featuring considerable touching components and mottled background. Aiming at identifying different regions in document images, la...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Penghai Zhao, Weilan Wang, Zhengqi Cai, Guowei Zhang, Yuqi Lu
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Document analysis and recognition fine-grained layout analysis historical Tibetan document images layout analysis text line segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/b0378dfd3723490a99c54a1ac5e0d762
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b0378dfd3723490a99c54a1ac5e0d762
record_format	dspace
spelling	oai:doaj.org-article:b0378dfd3723490a99c54a1ac5e0d7622021-11-25T00:00:53ZAccurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation2169-353610.1109/ACCESS.2021.3128536https://doaj.org/article/b0378dfd3723490a99c54a1ac5e0d7622021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9615155/https://doaj.org/toc/2169-3536Accurate layout analysis without subsequent text-line segmentation remains an ongoing challenge, especially when facing the Kangyur, a kind of historical Tibetan document featuring considerable touching components and mottled background. Aiming at identifying different regions in document images, layout analysis is indispensable for subsequent procedures such as character recognition. However, there was only a little research being carried out to perform line-level layout analysis which failed to deal with the Kangyur. To obtain the optimal results, a fine-grained sub-line level layout analysis approach is presented. Firstly, we introduced an accelerated method to build the dataset which is dynamic and reliable. Secondly, enhancement had been made to the SOLOv2 according to the characteristics of the Kangyur. Then, we fed the enhanced SOLOv2 with the prepared annotation file during the training phase. Once the network is trained, instances of the text line, sentence, and titles can be segmented and identified during the inference stage. The experimental results show that the proposed method delivers a decent 72.7% average precision on our dataset. In general, this preliminary research provides insights into the fine-grained sub-line level layout analysis and testifies the SOLOv2-based approaches. We also believe that the proposed methods can be adopted on other language documents with various layouts.Penghai ZhaoWeilan WangZhengqi CaiGuowei ZhangYuqi LuIEEEarticleDocument analysis and recognitionfine-grained layout analysishistorical Tibetan document imageslayout analysistext line segmentationElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 154435-154447 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Document analysis and recognition fine-grained layout analysis historical Tibetan document images layout analysis text line segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Document analysis and recognition fine-grained layout analysis historical Tibetan document images layout analysis text line segmentation Electrical engineering. Electronics. Nuclear engineering TK1-9971 Penghai Zhao Weilan Wang Zhengqi Cai Guowei Zhang Yuqi Lu Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
description	Accurate layout analysis without subsequent text-line segmentation remains an ongoing challenge, especially when facing the Kangyur, a kind of historical Tibetan document featuring considerable touching components and mottled background. Aiming at identifying different regions in document images, layout analysis is indispensable for subsequent procedures such as character recognition. However, there was only a little research being carried out to perform line-level layout analysis which failed to deal with the Kangyur. To obtain the optimal results, a fine-grained sub-line level layout analysis approach is presented. Firstly, we introduced an accelerated method to build the dataset which is dynamic and reliable. Secondly, enhancement had been made to the SOLOv2 according to the characteristics of the Kangyur. Then, we fed the enhanced SOLOv2 with the prepared annotation file during the training phase. Once the network is trained, instances of the text line, sentence, and titles can be segmented and identified during the inference stage. The experimental results show that the proposed method delivers a decent 72.7% average precision on our dataset. In general, this preliminary research provides insights into the fine-grained sub-line level layout analysis and testifies the SOLOv2-based approaches. We also believe that the proposed methods can be adopted on other language documents with various layouts.
format	article
author	Penghai Zhao Weilan Wang Zhengqi Cai Guowei Zhang Yuqi Lu
author_facet	Penghai Zhao Weilan Wang Zhengqi Cai Guowei Zhang Yuqi Lu
author_sort	Penghai Zhao
title	Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
title_short	Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
title_full	Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
title_fullStr	Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
title_full_unstemmed	Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation
title_sort	accurate fine-grained layout analysis for the historical tibetan document based on the instance segmentation
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/b0378dfd3723490a99c54a1ac5e0d762
work_keys_str_mv	AT penghaizhao accuratefinegrainedlayoutanalysisforthehistoricaltibetandocumentbasedontheinstancesegmentation AT weilanwang accuratefinegrainedlayoutanalysisforthehistoricaltibetandocumentbasedontheinstancesegmentation AT zhengqicai accuratefinegrainedlayoutanalysisforthehistoricaltibetandocumentbasedontheinstancesegmentation AT guoweizhang accuratefinegrainedlayoutanalysisforthehistoricaltibetandocumentbasedontheinstancesegmentation AT yuqilu accuratefinegrainedlayoutanalysisforthehistoricaltibetandocumentbasedontheinstancesegmentation
_version_	1718414713070551040

Accurate Fine-Grained Layout Analysis for the Historical Tibetan Document Based on the Instance Segmentation

Ejemplares similares