ABSTRACT
The segmentation of complex images into semantic regions has seen a growing interest these last years with the advent of Deep Learning. Until recently, most existing methods for Historical Document Analysis focused on the visual appearance of documents, ignoring the rich information that textual content can offer. However, the segmentation of complex documents into semantic regions is sometimes impossible relying only on visual features and recent models embed both visual and textual information. In this paper, we focus on the use of both visual and textual information for segmenting historical registers into structured and meaningful units such as acts. An act is a text recording containing valuable knowledge such as demographic information (baptism, marriage or death) or royal decisions (donation or pardon). We propose a simple pipeline to enrich document images with the position of text lines containing key-phrases and show that running a standard image-based layout analysis system on these images can lead to significant gains. Our experiments show that the detection of acts increases from 38 % of mAP to 74 % when adding textual information, in real use-case conditions where text lines positions and content are extracted with an automatic recognition system.
- Sofia Ares Oliveira, Benoit Seguin, and Frederic Kaplan. 2018. dhSegment: A generic deep-learning approach for document segmentation. In International Conference on Frontiers in Handwriting Recognition.Google ScholarCross Ref
- Ashish Arora, Chun Chieh Chang, Babak Rekabdar, Bagher BabaAli, Daniel Povey, David Etter, Desh Raj, Hossein Hadian, Jan Trmal, Paola Garcia, 2019. Using ASR methods for OCR. In International Conference on Document Analysis and Recognition.Google ScholarCross Ref
- Raphaël Barman, Maud Ehrmann, S. Clematide, S. Oliveira, and F. Kaplan. 2020. Combining Visual and Textual Features for Semantic Segmentation of Historical Newspapers. Journal of Data Mining & Digital Humanities, HistoInformatics, HistoInformatics (2020).Google Scholar
- Théodore Bluche, Sebastien Hamel, Christopher Kermorvant, Joan Puigcerver, Dominique Stutzmann, Alejandro H. Toselli, and Enrique Vidal. 2017. Preparatory KWS Experiments for Large-Scale Indexing of a Vast Medieval Manuscript Collection in the HIMANIS Project. In International Conference on Document Analysis and Recognition.Google ScholarCross Ref
- Mélodie Boillet, Christopher Kermorvant, and Thierry Paquet. 2020. Multiple Document Datasets Pre-training Improves Text Line Detection With Deep Neural Networks. In International Conference on Pattern Recognition.Google Scholar
- Mélodie Boillet, Christopher Kermorvant, and Thierry Paquet. 2021. Robust Text Line Detection in Historical Documents: Learning and Evaluation Methods. In Preparation.Google Scholar
- Tobias Grüning, Roger Labahn, Markus Diem, Florian Kleber, and Stefan Fiel. 2018. READ-BAD: A New Dataset and Evaluation Scheme for Baseline Detection in Archival Documents. In International Workshop on Document Analysis Systems.Google Scholar
- Tobias Grüning, Gundram Leifert, Tobias Strauß, and Roger Labahn. 2019. A Two-Stage Method for Text Line Detection in Historical Documents. International Journal on Document Analysis and Recognition (IJDAR) 22 (09 2019). https://doi.org/10.1007/s10032-019-00332-1Google ScholarDigital Library
- Paul Guérin and Léonce Celier. 1881-1958. Recueil des documents concernant le Poitou contenus dans les registres de la chancellerie de France. Société des archives historiques du Poitou, Poitiers.Google Scholar
- Olivier Guyotjeannin and Serge Lusignan. 2005. Le formulaire d’Odart Morchesne, dans la version du ms BNF fr. 5024. École des chartes, Paris.Google Scholar
- Philip Kahle, Sebastian Colutto, Günter Hackl, and Günter Mühlberger. 2017. Transkribus - A Service Platform for Transcription, Recognition and Retrieval of Historical Documents. In 2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR), Vol. 04. 19–24. https://doi.org/10.1109/ICDAR.2017.307Google Scholar
- O. Mechi, M. Mehri, R. Ingold, and N. Essoukri Ben Amara. 2019. Text Line Segmentation in Historical Document Images Using an Adaptive U-Net Architecture. In International Conference on Document Analysis and Recognition.Google Scholar
- José Ramón Prieto, Vicente Bosch, Enrique Vidal, Dominique Stutzmann, and Sébastien Hamel. 2020. Text Content Based Layout Analysis. In International Conference on Frontiers in Handwriting Recognition.Google Scholar
- Solène Tarride, Aurélie Lemaitre, Bertrand Coüasnon, and Sophie Tardivel. 2019. Signature Detection as a Way to Recognise Historical Parish Register Structure. In International Workshop on Historical Document Imaging and Processing.Google ScholarDigital Library
- Hélène Vézina and Jean-Sébastien Bournival. 2020. An Overview of the BALSAC Population Database. Past Developments, Current State and Future Prospects. In Historical Life Course Studies.Google Scholar
- Viard, Jules. 1899. Documents parisiens du règne de Philippe VI de Valois (1328-1350) : extraits des registres de la chancellerie de France. H. Champion, Paris.Google Scholar
- Gregor Wiedemann and Gerhard Heyer. 2018. Page Stream Segmentation with Convolutional Neural Nets Combining Textual and Visual Features. ArXiv abs/1710.03006(2018).Google Scholar
- Yiheng Xu, Minghao Li, Lei Cui, Shaohan Huang, Furu Wei, and Ming Zhou. 2020. LayoutLM: Pre-training of Text and Layout for Document Image Understanding. In International Conference on Knowledge Discovery & Data Mining.Google Scholar
- Xiao Yang, Ersin Yumer, Paul Asente, Mike Kraley, Daniel Kifer, and C. Lee Giles. 2017. Learning to Extract Semantic Structure from Documents Using Multimodal Fully Convolutional Neural Networks. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 4342–4351. https://doi.org/10.1109/CVPR.2017.462Google ScholarCross Ref
Recommendations
Text Retrieval for Japanese Historical Documents by Image Generation
HIP '17: Proceedings of the 4th International Workshop on Historical Document Imaging and ProcessingDigitization of historical documents is growing rapidly. Text retrieval is a vital technology to facilitate the use of historical document images because of the large amount of data. In this paper, we propose a method for retrieving keywords in Japanese ...
Automatic Corresponding Control Points Selection for Historical Document Image Registration
ICDAR '09: Proceedings of the 2009 10th International Conference on Document Analysis and RecognitionImage registration is crucial for various image analysis tasks. In particular, most approaches to correction of bleed-through distortion on handwritten document images require the recto image and the verso image to be precisely registered. In this paper,...
Keyword Analysis Visualization for Chinese Historical Texts
VINCI '19: Proceedings of the 12th International Symposium on Visual Information Communication and InteractionHistorical texts form the basis of the study of antiquities. In the case of Chinese historical texts different genres exist, e.g. chronological and biographical works etc. The contents of these texts normally consist of complex and interrelated ...
Comments