A Combined System for Text Line Extraction and Handwriting Recognition in Historical Documents

Author

Fischer, Anath ; Baechler, Micheal ; Garz, Angelika ; Liwicki, Marcus ; Ingold, Rolf

Author_Institution

Dept. of Electr. Eng., Polytech. Montreal, Montreal, QC, Canada

fYear

2014

fDate

7-10 April 2014

Firstpage

71

Lastpage

75

Abstract

Automated reading of historical handwriting is needed to search and browse ancient manuscripts in digital libraries based on their textual content. In this paper, we present a combined system for text localization and transcription in page images. It includes flexible learning-based methods for layout analysis and handwriting recognition, which were developed in the context of the Swiss research project HisDoc. A comprehensive experimental evaluation is provided for the medieval Parzival database, demonstrating a promising word recognition accuracy of 93.0% with closed vocabulary. In order to harmonize the evaluation of the two document analysis tasks, we introduce a novel evaluation measure for text line extraction that takes substitution, deletion, as well as insertion errors into account.

Keywords

digital libraries; document image processing; feature extraction; handwriting recognition; Swiss research project HisDoc; ancient manuscript; automated reading; digital libraries; document analysis task; flexible learning-based method; handwriting recognition; historical document; layout analysis; medieval Parzival database; text line extraction; text localization; transcription; word recognition; Accuracy; Databases; Handwriting recognition; Hidden Markov models; Layout; Text analysis; Text recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis Systems (DAS), 2014 11th IAPR International Workshop on

Conference_Location

Tours

Print_ISBN

978-1-4799-3243-6

Type

conf

DOI

10.1109/DAS.2014.51

Filename

6830972