A Component-Based Framework for Recognition Systems

  • Matthias Middendorf
  • Carsten Peust
  • Johannes Schacht
Conference paper

DOI: 10.1007/978-3-540-24642-8_10

Volume 2956 of the book series Lecture Notes in Computer Science (LNCS)
Cite this paper as:
Middendorf M., Peust C., Schacht J. (2004) A Component-Based Framework for Recognition Systems. In: Dengel A., Junker M., Weisbecker A. (eds) Reading and Learning. Lecture Notes in Computer Science, vol 2956. Springer, Berlin, Heidelberg

Abstract

The advantages of component-based software design have often been emphasized during the last years. This article describes the design and implementation of an object-oriented framework for document recognition systems. We use an operator-workitem-model for integrating numerous software components of various origins and written in different programming languages. This is done by defining two standardized interfaces, namely an operator-interface for the integrated recognition algorithms and a workitem-interface for all in- and output objects of the recognition algorithms. These conditions allow to arrange the operators in any successive order and any complexity to build a complete recognition system. For any choice and relative position of operators integrated, the framework provides facilities for parametrizing, testing and running.

The reported works have been part of the project AdaptiveREAD which was funded by the German Federal Ministry of Education and Research.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Matthias Middendorf
    • 1
  • Carsten Peust
    • 1
  • Johannes Schacht
    • 1
  1. 1.Océ Document Technologies GmbHKonstanzGermany