Europeana Newspapers OCR Workflow Evaluation
Authors:
Open Access
- 22 August 2015
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 39-46
Abstract
This paper summarises the final performance evaluation results of the OCR workflow which was employed for large-scale production in the Europeana Newspapers project. It gives a detailed overview of how the involved software performed on a representative dataset of historical newspaper pages (for which ground truth was created) with regard to general text accuracy as well as layout-related factors which have an impact on how the material can be used in specific use scenarios. Specific types of errors are examined and evaluated in order to identify possible improvements related to the employed document image analysis and recognition methods. Moreover, alternatives to the standard production workflow are assessed to determine future directions and give advice on best practice related to OCR projects.Keywords
Funding Information
- European Research Council (297380)
This publication has 9 references indexed in Scilit:
- The ENP image and ground truth dataset of historical newspapers2015
- ICDAR2015 competition on recognition of documents with complex layouts - RDCL20152015
- The Significance of Reading Order in Document Recognition and Its Evaluation2013
- ICDAR 2013 Competition on Historical Newspaper Layout Analysis (HNLA 2013)2013
- ICDAR 2013 Competition on Historical Book Recognition (HBR 2013)2013
- Scenario Driven In-depth Performance Evaluation of Document Layout Analysis Methods2011
- The PAGE (Page Analysis and Ground-Truth Elements) Format Framework2010
- Measuring Mass Text Digitization Quality and UsefulnessD-Lib Magazine, 2009
- How Good Can It Get?D-Lib Magazine, 2009