A large-scale dataset for Chinese historical document recognition and analysis
Abstract The development of Chinese civilization has produced a vast collection of historical documents. Recognizing and analyzing these documents hold significant value for the research of ancient culture. Recently, researchers have tried to utilize deep-learning techniques to automate recognition...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Data |
| Online Access: | https://doi.org/10.1038/s41597-025-04495-x |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571981836320768 |
|---|---|
| author | Yongxin Shi Dezhi Peng Yuyi Zhang Jiahuan Cao Lianwen Jin |
| author_facet | Yongxin Shi Dezhi Peng Yuyi Zhang Jiahuan Cao Lianwen Jin |
| author_sort | Yongxin Shi |
| collection | DOAJ |
| description | Abstract The development of Chinese civilization has produced a vast collection of historical documents. Recognizing and analyzing these documents hold significant value for the research of ancient culture. Recently, researchers have tried to utilize deep-learning techniques to automate recognition and analysis. However, existing Chinese historical document datasets, which are heavily relied upon by deep-learning models, suffer from limited data scale, insufficient character category, and lack of book-level annotation. To fill this gap, we introduce HisDoc1B, a large-scale dataset for Chinese historical document recognition and analysis. The HisDoc1B comprises 40,281 books, over 3 million document images, and over 1 billion characters across 30,615 character categories. To the best of our knowledge, HisDoc1B is the largest dataset in the field, surpassing existing datasets by more than 200 times in scale. Additionally, it is the only dataset with book-level annotations and punctuation annotations. Furthermore, extensive experiments demonstrate the high quality and practical utility of the proposed HisDoc1B. We believe that HisDoc1B could provide valuable resources to boost the advancement of research in this domain. |
| format | Article |
| id | doaj-art-17ff5ca0cbd145e587be8982110ae7be |
| institution | Kabale University |
| issn | 2052-4463 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Data |
| spelling | doaj-art-17ff5ca0cbd145e587be8982110ae7be2025-02-02T12:08:24ZengNature PortfolioScientific Data2052-44632025-01-0112111010.1038/s41597-025-04495-xA large-scale dataset for Chinese historical document recognition and analysisYongxin Shi0Dezhi Peng1Yuyi Zhang2Jiahuan Cao3Lianwen Jin4School of Electronic and Information Engineering, South China University of TechnologySchool of Electronic and Information Engineering, South China University of TechnologySchool of Electronic and Information Engineering, South China University of TechnologySchool of Electronic and Information Engineering, South China University of TechnologySchool of Electronic and Information Engineering, South China University of TechnologyAbstract The development of Chinese civilization has produced a vast collection of historical documents. Recognizing and analyzing these documents hold significant value for the research of ancient culture. Recently, researchers have tried to utilize deep-learning techniques to automate recognition and analysis. However, existing Chinese historical document datasets, which are heavily relied upon by deep-learning models, suffer from limited data scale, insufficient character category, and lack of book-level annotation. To fill this gap, we introduce HisDoc1B, a large-scale dataset for Chinese historical document recognition and analysis. The HisDoc1B comprises 40,281 books, over 3 million document images, and over 1 billion characters across 30,615 character categories. To the best of our knowledge, HisDoc1B is the largest dataset in the field, surpassing existing datasets by more than 200 times in scale. Additionally, it is the only dataset with book-level annotations and punctuation annotations. Furthermore, extensive experiments demonstrate the high quality and practical utility of the proposed HisDoc1B. We believe that HisDoc1B could provide valuable resources to boost the advancement of research in this domain.https://doi.org/10.1038/s41597-025-04495-x |
| spellingShingle | Yongxin Shi Dezhi Peng Yuyi Zhang Jiahuan Cao Lianwen Jin A large-scale dataset for Chinese historical document recognition and analysis Scientific Data |
| title | A large-scale dataset for Chinese historical document recognition and analysis |
| title_full | A large-scale dataset for Chinese historical document recognition and analysis |
| title_fullStr | A large-scale dataset for Chinese historical document recognition and analysis |
| title_full_unstemmed | A large-scale dataset for Chinese historical document recognition and analysis |
| title_short | A large-scale dataset for Chinese historical document recognition and analysis |
| title_sort | large scale dataset for chinese historical document recognition and analysis |
| url | https://doi.org/10.1038/s41597-025-04495-x |
| work_keys_str_mv | AT yongxinshi alargescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT dezhipeng alargescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT yuyizhang alargescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT jiahuancao alargescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT lianwenjin alargescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT yongxinshi largescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT dezhipeng largescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT yuyizhang largescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT jiahuancao largescaledatasetforchinesehistoricaldocumentrecognitionandanalysis AT lianwenjin largescaledatasetforchinesehistoricaldocumentrecognitionandanalysis |