Skip to main content
Springer Nature Link
Log in

Adaptive Combination of Commercial OCR Systems

  • Chapter
Reading and Learning

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 2956))

  • 432 Accesses

Abstract

Combining multiple classifiers to achieve improved recognition results has become a popular technique in recent years. As for OCR systems, most investigations focus on fusion strategies on the character level. This paper describes a flexible framework for the combination of result strings which are the common output of commercial OCR systems. By synchronizing strings according to geometrical criteria, incorrect character segmentations can be avoided, while character recognition is improved by classical combination rules like Borda Count or Plurality Vote. To reduce computing time, further expert calls are stopped as soon as the quality of a temporary combination result exceeds a given threshold. The system allows easy integration of arbitrary new OCR systems and simplifies the determination of optimal system parameters by analyzing the input data at hand. Quantitative results are shown for a two-recognizer system, while the framework allows an arbitrary number of experts.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Aksela, M.: Comparison of Classifier Selection Methods for Improving Committee Performance. In: [WiRo 03], pp. 84–93

    Google Scholar 

  2. Alexandre, L., Campilho, A., Kamel, M.: On Combining Classifiers Using Sum and Product Rules. Pattern Recognition Letters 22, 1283–1289 (2001)

    Article  MATH  Google Scholar 

  3. Alkoot, F.M., Kittler, J.: Experimental Evaluation of Expert Fusion Strategies. Pattern Recognition Letters 20, 1361–1369 (1999)

    Article  Google Scholar 

  4. Altinçay, H., Demirekler, M.: Post-processing of Classifier Outputs in Multiple Classifier Systems. In: [RoKi02], pp. 159–168

    Google Scholar 

  5. Bauer, E., Kohavi, R.: An Empirical Comparison of Voting Classification Algorithms: Bagging, Boosting, and Variants. Machine Learning 36, 105–139 (1999)

    Article  Google Scholar 

  6. Bouchaffra, D., Govindaraju, V., Srihari, S.: A Methodology for Mapping Scores to Probabilities. IEEE Trans. Pattern Analysis and Machine Intelligence 21, 923–927 (1999)

    Article  Google Scholar 

  7. Constantinidis, A., Fairhurst, M., Rahman, A.: A New Multi-Expert Decision Combination Algorithm and its Application to the Detection of Circumscribed Masses in Digital Mammogramms. Pattern Recognition 34, 1527–1537 (2001)

    Article  MATH  Google Scholar 

  8. Dietterich, T.G.: Ensemble Methods in Machine Learning. In: [KiRo00], pp. 1-15

    Google Scholar 

  9. Duin, R., Tax, D.: Experiments with Classifier Combining Rules. In: [KiRo00], pp. 16-29

    Google Scholar 

  10. van Erp, M., Schomaker, L.: Variants of the Borda Count Method for combining ranked classifier hypotheses. In: Proceedings FIHR 2000, pp. 443 - 452 (2000)

    Google Scholar 

  11. Fumera, G., Roli, F.: Linear Combination for Classifier Fusion: Some Theoretical and Experimental Results. In: [WiRo03], pp. 74-83

    Google Scholar 

  12. Giacinto, G., Roli, F.: Dynamic Classifier Selection Based on Multiple Classifier Behaviour. Pattern Recognition 34, 1879–1881 (2001)

    Article  MATH  Google Scholar 

  13. Günter, S., Bunke, H.: New Boosting Algorithms for Classification Problems with Large Number of Classes Applied to Handwritten Word Recognition Task. In [WiRo03], pp. 326-335

    Google Scholar 

  14. Ho, T.K.: The Random Subspace Method for Constructing Decision Forests. IEEE Trans. Pattern Analysis and Machine Intelligence 20, 832–844 (1998)

    Article  Google Scholar 

  15. Kamel, M., Wanas, N., Data Dependence in Combining Classifiers. In: [WIRo03], pp. 1-14

    Google Scholar 

  16. Kittler, J., Roli, F. (eds.): Multiple Classifier Systems 2000. LNCS, vol. 1857. Springer, Heidelberg (2000)

    Google Scholar 

  17. Kittler, J., Roli, F. (eds.): Multiple Classifier System 2001. LNCS, vol. 2096. Springer, Heidelberg (2001)

    MATH  Google Scholar 

  18. Klink, S.: Entwurf, Implementierung und Vergleich von Algorithmen zum Merge von Segmentierungs- und Klassifikationsergebnissen unterschiedlicher OCRErgebnisse. Master Thesis, Kaiserslautern (1997)

    Google Scholar 

  19. Kuncheva, L., Bezdek, J., Duin, R.: Decision Templates for Multiple Classifier Fusion: an Experimental Comparison. Pattern Recognition 34, 299–314

    Google Scholar 

  20. Lam, L., Huang, Y.-S., Suen, C.: Combination of Multiple Classifier Decisions for Optical Character Recognition. In: Bunke, H., Wang, P.S.P. (eds.) Handbook of Character Recognition and Document Image Analysis, pp. 79–101. World Scientific, Singapore (1997)

    Google Scholar 

  21. Lin, X., Ding, X., Chen, M., Zhang, R., Wu, Y.: Adaptive Confidence Transform Based Classifier Combination for Chinese Character Recognition. Pattern Recognition Letters 19, 957–988 (1998)

    Google Scholar 

  22. Lin, X., Yacoub, S., Burns, J., Simske, S.: Performance Analysis of Pattern Classifier Combination by Plurality Voting. Pattern Recognition Letters 24, 1959–1969 (2003)

    Article  Google Scholar 

  23. Marti, U.-V., Bunke, H.: Use of Positional Information in Sequence Alignment for Multiple Classifier Information. In [KiRo01] pp. 388-398

    Google Scholar 

  24. Masulli, F., Valentini, G.: Effectiveness of Error Correcting Output Coding Methods in Ensemble and Monolithic Learning Machines, Preprint, Università di Genova (2002)

    Google Scholar 

  25. Raudys, Š., Roli, F.: The Behavior Knowledge Space Fusion Method: Analysis of Generalization Error and Strategies for Performance Improvement. In: [RaR003], pp. 55-64

    Google Scholar 

  26. Kittler, J., Roli, F. (eds.): Multiple Classifier Systems 2002. LNCS, vol. 2364. Springer, Heidelberg (2002)

    MATH  Google Scholar 

  27. Ruta, D., Gabrys, B.: New Measure of Classifier Dependency in Multiple Classifier Systems. In: [RoKi02], pp. 127-136

    Google Scholar 

  28. Skurichina, M., Duin, R.: Bagging and the Random Subspace Method for Redundant Feature Spaces. In: [KiRo01], pp. 1-10

    Google Scholar 

  29. Sirlantzis, K., Hoque, S., Fairhurst, M.: Trainable Multiple Classifier Schemes for Handwritten Character Recognition. In: [RoKi02], pp. 169-178

    Google Scholar 

  30. Suen, Ch., Lam, L., Multiple Classifier Combination Methodologies for Different Output Levels. In: [KiRo00], pp. 52-66

    Google Scholar 

  31. Tax, D., van Breukelen, M., Duin, R., Kittler, J.: Combining Multiple Classifiers by Averaging or by Multiplying? Pattern Recognition 33, 1475–1485 (2000)

    Article  Google Scholar 

  32. Verma, B., Gader, P., Chen, W.: Fusion of Multiple Handwritten Word Recognition Techniques. Pattern Recognition Letters 22, 991–998 (2001)

    Article  MATH  Google Scholar 

  33. Wang, W., Brakensiek, A., Rigoll, G.: Combination of multiple classifiers for handwritten word recognition. In: 8th Int. Workshop on Frontiers in Handwriting Recognition, pp. 117-122 (2002)

    Google Scholar 

  34. Wang, J., Yan, H.: A Hybrid Method for Unconstrained Handwritten Numeral Recognition by Combining Structural and Neural Gas Classifiers. Pattern Recognition Letters 21, 625–635 (2000)

    Article  Google Scholar 

  35. Wilczok, L., Lellmann, W.: Design and Evaluation of an Adaptive Combination Framework for OCR Result Strings. In: [WiRo03], pp. 395-404

    Google Scholar 

  36. Windeatt, T., Roli, F. (eds.): Multiple Classifier Systems 2003. LNCS, vol. 2709. Springer, Heidelberg (2003)

    MATH  Google Scholar 

  37. Xu, L., Krzyżak, A., Suen, C.Y.: Methods of Combining Multiple Classifiers and Their Applications to Handwriting Recognition. IEEE Transactions on Systems, Man and Cybernetics 22, 418–435 (1992)

    Article  Google Scholar 

  38. Ye, X., Cheriet, M., Suen, C.Y.: StrCombo: combination of string recognizers. Pattern Recognition Letters 23, 381–394 (2002)

    Article  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Océ Document Technologies GmbH, Constance, Germany

    Elke Wilczok & Wolfgang Lellmann

Authors
  1. Elke Wilczok
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wolfgang Lellmann
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Editors and Affiliations

  1. Knowledge Management Department, German Research Center for Artificial Intelligence (DFKI) GmbH, Kaiserslautern, Germany

    Andreas Dengel

  2. German Research Center for Artificial Intelligence (DFKI), P.O.Box 2080, 67608, Kaiserslautern, Germany

    Markus Junker

  3. Institute for Human Factors and Technology Management, University of Stuttgart, Nobelstraße 12, 70569, Stuttgart, Germany

    Anette Weisbecker

Copyright information

© 2004 Springer-Verlag Berlin Heidelberg

About this chapter

Cite this chapter

Wilczok, E., Lellmann, W. (2004). Adaptive Combination of Commercial OCR Systems. In: Dengel, A., Junker, M., Weisbecker, A. (eds) Reading and Learning. Lecture Notes in Computer Science, vol 2956. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-24642-8_8

Download citation

  • DOI: https://doi.org/10.1007/978-3-540-24642-8_8

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-21904-0

  • Online ISBN: 978-3-540-24642-8

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics