Historical Documents and automatic text recognition

With this special issue of the Journal of Data Mining and Digital Humanities (JDMDH), we wish to bring together in one single volume several experiments, projects and reflections related to automatic text recognition on Historical documents. This special issue is the outcome of an event that took place at the Ecole Nationale des Chartes in Paris on June 23 and 24, 2022, which brought together scholars from various backgrounds to discuss the use of HTR and OCR in their researches. During these days, problems of engineering, machine learning or infrastructure were raised. Many technical subjects such as segmentation or the development of models linked to philological questions were discussed. The different speeches covered a large number of documents: manuscripts, archives, epigraphic materials, documents, sometimes in languages with their own specificities such as Hebrew, CHAM or ancient Greek from the 11th to the 20th century.


1. You Actually Look Twice At it (YALTAi): using an object detection approach instead of region segmentation within the Kraken engine

Thibault Clérice.
Layout Analysis (the identification of zones and their classification) is the first step along line segmentation in Optical Character Recognition and similar tasks. The ability of identifying main body of text from marginal text or running titles makes the difference between extracting the work full text of a digitized book and noisy outputs. We show that most segmenters focus on pixel classification and that polygonization of this output has not been used as a target for the latest competition on historical document (ICDAR 2017 and onwards), despite being the focus in the early 2010s. We propose to shift, for efficiency, the task from a pixel classification-based polygonization to an object detection using isothetic rectangles. We compare the output of Kraken and YOLOv5 in terms of segmentation and show that the later severely outperforms the first on small datasets (1110 samples and below). We release two datasets for training and evaluation on historical documents as well as a new package, YALTAi, which injects YOLOv5 in the segmentation pipeline of Kraken 4.1.

2. Generic HTR Models for Medieval Manuscripts. The CREMMALab Project

Ariane Pinche.
In the Humanities, the emergence of digital methods has opened up research questions to quantitative analysis. This is why HTR technology is increasingly involved in humanities research projects following precursors such as the Himanis project. However, many research teams have limited resources, either financially or in terms of their expertise in artificial intelligence. It may therefore be difficult to integrate handwritten text recognition into their project pipeline if they need to train a model or to create data from scratch. The goal here is not to explain how to build or improve a new HTR engine, nor to find a way to automatically align a preexisting corpus with an image to quickly create ground truths for training. This paper aims to help humanists easily develop an HTR model for medieval manuscripts, create and gather training data by knowing the issues underlying their choices. The objective is also to show the importance of the constitution of consistent data as a prerequisite to allow their gathering and to train efficient HTR models. We will present an overview of our work and experiment in the CREMMALab project (2021-2022), showing first how we ensure the consistency of the data and then how we have developed a generic model for medieval French manuscripts from the 13 th to the 15 th century, ready to be shared (more than 94% accuracy) and/or fine-tuned by other projects.

3. Towards a general open dataset and model for late medieval Castilian text recognition (HTR/OCR)

Matthias Gille Levenson.
Submitted to the Journal of Data Mining and Digital Humanities, and accepted. Pending last revisions. Please cite: @article{gille_levenson_2023_towards, author = {Gille Levenson, Matthias}, date = {2023}, journaltitle = {Journal of Data Mining and Digital Humanities}, doi = {10.5281/zenodo.7387376}, editor = {Pinche, Ariane and Stokes, Peter}, issuetitle = {Special Issue: Historical documents and automatic text recognition}, title = {Towards a general open dataset and models for late medieval Castilian text recognition (HTR/OCR)}, note = {Accepted, to be published.} } GILLE LEVENSON , Matthias, « Towards a general open dataset and models for late medieval Castilian text recognition (HTR/OCR) », Journal of Data Mining and Digital Humanities (2023) : Special Issue : Historical documents and automatic text recognition, eds. Ariane PINCHE and Peter STOKES, DOI : 10.5281/zenodo.7387376. Link to the data: https://doi.org/10.5281/zenodo.7386489

4. Impact of Image Enhancement Methods on Automatic Transcription Trainings with eScriptorium

Pauline Jacsont ; Elina Leblanc.
This study stems from the Desenrollando el cordel (Untangling the cordel) project, which focuses on 19th-century Spanish prints editing. It evaluates the impact of image enhancement methods on the automatic transcription of low-quality documents, both in terms of printing and digitisation. We compare different methods (binarisation, deblur) and present the results obtained during the training of models with the Kraken tool. We demonstrate that binarisation methods give better results than the other, and that the combination of several techniques did not significantly improve the transcription prediction. This study shows the significance of using image enhancement methods with Kraken. It paves the way for further experiments with larger and more varied corpora to help future projects design their automatic transcription workflow.

5. EpiSearch. Identifying Ancient Inscriptions in Epigraphic Manuscripts

Lorenzo Calvelli ; Federico Boschetti ; Tatiana Tommasi.
AbstractEpigraphic documents are an essential source of evidence for our knowledge of the ancient world. Nonetheless, a significant number of inscriptions have not been preserved in their material form. In fact, their texts can only be recovered thanks to handwritten materials and, in particular, the so-called epigraphic manuscripts. EpiSearch is a pilot project that explores the application of digital technologies deployed to retrieve the epigraphic evidence found in these sources. The application of Handwritten Text Recognition (HTR) to epigraphic manuscripts is a challenging task, given the nature and graphic layout of these documents. Yet, our research shows that, even with some limits, HTR technologies can be used successfully. 
Section: Sciences of Antiquity and digital humanities

6. Handwritten Text Recognition for Documentary Medieval Manuscripts

Sergio Torres Aguilar ; Vincent Jolivet.
Handwritten Text Recognition (HTR) techniques aim to accurately recognize sequences of characters in input manuscript images by training artificial intelligence models to capture historical writing features. Efficient HTR models can transform digitized manuscript collections into indexed and quotable corpora, providing valuable research insight for various historical inquiries. However, several challenges must be addressed, including the scarcity of relevant training corpora, the consequential variability introduced by different scribal hands and writing scripts, and the complexity of page layouts. This paper presents two models and one cross-model approach for automatic transcription of Latin and French medieval documentary manuscripts, particularly charters and registers, written between the 12th and 15th centuries and classified into two major writing scripts: Textualis (from the late-11th to 13th century) and Cursiva (from the 13th to the 15th century). The architecture of the models is based on a Convolutional Recurrent Neural Network (CRNN) coupled with a Connectionist Temporal Classification (CTC) loss. The training and evaluation of the models, involving 120k lines of text and almost 1M tokens, were conducted using three available ground-truth corpora : The e-NDP corpus, the Alcar-HOME database and the Himanis project. This paper describes the training architecture and corpora used, while discussing the main training challenges, results, and potential applications of […]

7. Toward Automatic Typography Analysis: Serif Classification and Font Similarities

Syed Talal Wasim ; Romain Collaud ; Lara Défayes ; Nicolas Henchoz ; Mathieu Salzmann ; Delphine Ribes Lemay.
Whether a document is of historical or contemporary significance, typography plays a crucial role in its composition. From the early days of modern printing, typographic techniques have evolved and transformed, resulting in changes to the features of typography. By analyzing these features, we can gain insights into specific time periods, geographical locations, and messages conveyed through typography. Therefore, in this paper, we aim to investigate the feasibility of training a model to classify serif typeswithout knowledge of the font and character. We also investigate how to train a vectorial-based image model able to group together fonts with similar features. Specifically, we compare the use of state-of-theart image classification methods, such as the EfficientNet-B2 and the Vision Transformer Base model with different patch sizes, and the state-of-the-art fine-grained image classification method, TransFG, on the serif classification task. We also evaluate the use of the DeepSVG model to learn to group fonts with similar features. Our investigation reveals that fine-grained image classification methods are better suited for the serif classification tasks and that leveraging the character labels helps to learn more meaningful font similarities.This repository contains: - Paper published in the Journal of data mining and digital humanities:WasimEtAl_Toward_Automatic_Typography_Analysis__Serif_Classification_and_Font_Similarities.pdf - Two datasets: The first […]
Section: Project presentations

8. Preparing Big Manuscript Data for Hierarchical Clustering with Minimal HTR Training

Elpida Perdiki.
HTR (Handwritten Text Recognition) technologies have progressed enough to offer high-accuracy results in recognising handwritten documents, even on a synchronous level. Despite the state-of-the-art algorithms and software, historical documents (especially those written in Greek) remain a real-world challenge for researchers. A large number of unedited or under-edited works of Greek Literature (ancient or Byzantine, especially the latter) exist to this day due to the complexity of producing critical editions. To critically edit a literary text, scholars need to pinpoint text variations on several manuscripts, which requires fully (or at least partially) transcribed manuscripts. For a large manuscript tradition (i.e., a large number of manuscripts transmitting the same work), such a process can be a painstaking and time-consuming project. To that end, HTR algorithms that train AI models can significantly assist, even when not resulting in entirely accurate transcriptions. Deep learning models, though, require a quantum of data to be effective. This, in turn, intensifies the same problem: big (transcribed) data require heavy loads of manual transcriptions as training sets. In the absence of such transcriptions, this study experiments with training sets of various sizes to determine the minimum amount of manual transcription needed to produce usable results. HTR models are trained through the Transkribus platform on manuscripts from multiple works of a single Byzantine author, […]
Section: Sciences of Antiquity and digital humanities

9. The Challenges of HTR Model Training: Feedback from the Project Donner le gout de l'archive a l'ere numerique

Beatrice Couture ; Farah Verret ; Maxime Gohier ; Dominique Deslandres.
The arrival of handwriting recognition technologies offers new possibilities for research in heritage studies. However, it is now necessary to reflect on the experiences and the practices developed by research teams. Our use of the Transkribus platform since 2018 has led us to search for the most significant ways to improve the performance of our handwritten text recognition (HTR) models which are made to transcribe French handwriting dating from the 17th century. This article therefore reports on the impacts of creating transcribing protocols, using the language model at full scale and determining the best way to use base models in order to help increase the performance of HTR models. Combining all of these elements can indeed increase the performance of a single model by more than 20% (reaching a Character Error Rate below 5%). This article also discusses some challenges regarding the collaborative nature of HTR platforms such as Transkribus and the way researchers can share their data generated in the process of creating or training handwritten text recognition models.

10. Incorporating Crowdsourced Annotator Distributions into Ensemble Modeling to Improve Classification Trustworthiness for Ancient Greek Papyri

Graham West ; Matthew I. Swindall ; Ben Keener ; Timothy Player ; Alex C. Williams ; James H. Brusuelas ; John F. Wallin.
Performing classification on noisy, crowdsourced image datasets can prove challenging even for the best neural networks. Two issues which complicate the problem on such datasets are class imbalance and ground-truth uncertainty in labeling. The AL-ALL and AL-PUB datasets - consisting of tightly cropped, individual characters from images of ancient Greek papyri - are strongly affected by both issues. The application of ensemble modeling to such datasets can help identify images where the ground-truth is questionable and quantify the trustworthiness of those samples. As such, we apply stacked generalization consisting of nearly identical ResNets with different loss functions: one utilizing sparse cross-entropy (CXE) and the other Kullback-Liebler Divergence (KLD). Both networks use labels drawn from a crowd-sourced consensus. This consensus is derived from a Normalized Distribution of Annotations (NDA) based on all annotations for a given character in the dataset. For the second network, the KLD is calculated with respect to the NDA. For our ensemble model, we apply a k-nearest neighbors model to the outputs of the CXE and KLD networks. Individually, the ResNet models have approximately 93% accuracy, while the ensemble model achieves an accuracy of > 95%, increasing the classification trustworthiness. We also perform an analysis of the Shannon entropy of the various models' output distributions to measure classification uncertainty. Our results suggest that entropy is […]
Section: Digital humanities in languages

11. Exploring Data Provenance in Handwritten Text Recognition Infrastructure: Sharing and Reusing Ground Truth Data, Referencing Models, and Acknowledging Contributions. Starting the Conversation on How We Could Get It Done

C. Annemieke Romein ; Tobias Hodel ; Femke Gordijn ; Joris J. van Zundert ; Alix Chagué ; Milan van Lange ; Helle Strandgaard Jensen ; Andy Stauder ; Jake Purcell ; Melissa M. Terras et al.
This paper discusses best practices for sharing and reusing Ground Truth in Handwritten Text Recognition infrastructures, as well as ways to reference and acknowledge contributions to the creation and enrichment of data within these systems. We discuss how one can place Ground Truth data in a repository and, subsequently, inform others through HTR-United. Furthermore, we want to suggest appropriate citation methods for ATR data, models, and contributions made by volunteers. Moreover, when using digitised sources (digital facsimiles), it becomes increasingly important to distinguish between the physical object and the digital collection. These topics all relate to the proper acknowledgement of labour put into digitising, transcribing, and sharing Ground Truth HTR data. This also points to broader issues surrounding the use of machine learning in archival and library contexts, and how the community should begin to acknowledge and record both contributions and data provenance.

12. Historical Documents and Automatic Text Recognition: Introduction

Ariane Pinche ; Peter Stokes.
With this special issue of the Journal of Data Mining and Digital Humanities (JDMDH), we bringtogether in one single volume several experiments, projects and reflections related to automatic textrecognition applied to historical documents.More and more research projects1 now include automatic text acquisition in their data processing chain,and this is true not only for projects focussed on Digital or Computational Humanities but increasinglyalso for those that are simply using existing digital tools as the means to an end. The increasing useof this technology has led to an automation of tasks that affects the role of the researcher in the textualproduction process. This new data-intensive practice makes it urgent to collect and harmonise the corporanecessary for the constitution of training sets, but also to make them available for exploitation. Thisspecial issue is therefore an opportunity to present articles combining philological and technical questionsto make a scientific assessment of the use of automatic text recognition for ancient documents, itsresults, its contributions and the new practices induced by its use in the process of editing and exploringtexts. We hope that practical aspects will be questioned on this occasion, while raising methodologicalchallenges and its impact on research data.The special issue on Automatic Text Recognition (ATR) is therefore dedicated to providing a comprehensiveoverview of the use of ATR in the humanities field, particularly […]