OCR4all—An Open-Source Tool Providing a (Semi-)Automatic OCR Workflow for Historical Printings

Christian Reul,Christoph Wick,Frank Puppe,Nico Balbach,Christine Grundig,Uwe Springmann,Alexander Hartelt,Maximilian Wehner,Dennis Christ,Andreas Büttner

doi:10.3390/app9224853

Abstract

Optical Character Recognition (OCR) on historical printings is a challenging task mainly due to the complexity of the layout and the highly variant typography. Nevertheless, in the last few years, great progress has been made in the area of historical OCR, resulting in several powerful open-source tools for preprocessing, layout analysis and segmentation, character recognition, and post-processing. The drawback of these tools often is their limited applicability by non-technical users like humanist scholars and in particular the combined use of several tools in a workflow. In this paper, we present an open-source OCR software called OCR4all, which combines state-of-the-art OCR components and continuous model training into a comprehensive workflow. While a variety of materials can already be processed fully automatically, books with more complex layouts require manual intervention by the users. This is mostly due to the fact that the required ground truth for training stronger mixed models (for segmentation, as well as text recognition) is not available, yet, neither in the desired quantity nor quality. To deal with this issue in the short run, OCR4all offers a comfortable GUI that allows error corrections not only in the final output, but already in early stages to minimize error propagations. In the long run, this constant manual correction produces large quantities of valuable, high quality training material, which can be used to improve fully automatic approaches. Further on, extensive configuration capabilities are provided to set the degree of automation of the workflow and to make adaptations to the carefully selected default parameters for specific printings, if necessary. During experiments, the fully automated application on 19th Century novels showed that OCR4all can considerably outperform the commercial state-of-the-art tool ABBYY Finereader on moderate layouts if suitably pretrained mixed OCR models are available. Furthermore, on very complex early printed books, even users with minimal or no experience were able to capture the text with manageable effort and great quality, achieving excellent Character Error Rates (CERs) below 0.5%. The architecture of OCR4all allows the easy integration (or substitution) of newly developed tools for its main components by standardized interfaces like PageXML, thus aiming at continual higher automation for historical printings.

Highlights

While Optical Character Recognition (OCR) is regularly considered to be a solved problem [1], gathering the textual content of historical printings using OCR can still be a very challenging and cumbersome task [2], due to various reasons
We rate the engines mainly based on their speed and effectiveness, and take into account the user friendliness when it comes to training on real data
To keep the manual effort to a minimum, we introduced an Iterative Training Approach (ITA), which is fully supported by OCR4all

Summary

Introduction

While Optical Character Recognition (OCR) is regularly considered to be a solved problem [1], gathering the textual content of historical printings using OCR can still be a very challenging and cumbersome task [2], due to various reasons. Among the problems that need to be addressed for early printings is the often intricate layout containing images, ornaments, marginal notes, and swash capitals. While modern fonts can be recognized with excellent accuracy by so-called omnifont or polyfont models, very early printings like incunabula (books printed before 1501), and handwritten texts usually require book-specific training in order to reach Character Error Rates (CERs) well below. In the last few years, some progress has been made in the area of historical OCR, especially concerning the character recognition problem. An important milestone was the introduction of recurrent neural networks with Long Short Term Memory (LSTM) [5] trained using a Connectionist

Objectives

Methods

Results

Discussion

Conclusion