Summary
In the paper the multilevel probabilistic approach to hand printed form recognition is described. The form recognition is decomposed into two levels: character recognition and word recognition. On the letter level the rough sets approach is presented. After this level of classification, for every position in the word, we obtain either the certain or the subset of possible or the subset of impossible decision about recognized letter. After on the word level the probabilistic lexicons are available. The decision on the word level is performed using probabilistic properties of character classifier and the contents of probabilistic lexicon. The novel approach to combining these two sources of information about classes (words) probabilities is proposed, which is based on lexicons and accuracy assessment of local character classifiers. Some experimental results and examples of practical applications of recognition method are also briefly described.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Fibak, J., Pawlak, Z., Slowinski, K., Slowinski, R.: Rough Set Based Decision Algorithm for Treatment of Duodenal Ulcer by HSV. Bull. of the Polish Acad. Sci., Bio Sci. 34, 227–246 (1986)
Grzymala-Busse, J.: A System for Learning from Examples Based on Rough Sets. In: Slowinski, R. (ed.) Intelligent Decision Support: Handbook of Applications and Advances of the Rough Sets Theory, pp. 3–18. Kluwer Academic Publishers, Dordrecht (1992)
Kuncheva, L.: Combining Classifiers: Soft Computing Solutions. In: Pal, S., Pal, A. (eds.) Pattern Recognition: from Classical to Modern Approaches, pp. 427–451. World Scientific, Singapore (2001)
Liu, C., Nakashima, K., Sako, H.: Handwritten Digit Recognition: Benchmarking of State-of-the-Art Techniques. Pattern Recognition 36, 2271–2285 (2003)
Marti, U.V., Bunke, H.: Using a Statistical Language Model to Improve the Performance of an HMM-Based Cursive Handwritting Recognition System. Int. Journ. of Pattern Recognition and Artificial Intelligence 15, 65–90 (2001)
Pawlak, Z.: Rough Sets, Decision Algorithms and Bayes Theorem. European Journal of Operational Research 136, 181–189 (2002)
Sas, J., Kurzynski, M.: Multilevel Recognition of Structured Handwritten Documents - Probabilistic Approach. In: Proc. 4th Int. Conf. on Computer Recognition Systems, pp. 723–730. Springer, Heidelberg (2005)
Sas, J., Kurzynski, M.: Combining Character Level Classifier and Probabilistic Lexicons in Handwritten Word Recognition - Comparative Analysis of Methods. In: Gagalowicz, A., Philips, W. (eds.) CAIP 2005. LNCS, vol. 3691, pp. 330–337. Springer, Heidelberg (2005)
Sas, J., Zolnierek, A.: Comparison of feature reduction methods in the text recognition task. In: Rutkowski, L., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2008. LNCS(LNAI), vol. 5097, pp. 729–738. Springer, Heidelberg (2008)
Zolnierek, A.: Application of rough sets theory to the sequential diagnosis. In: Maglaveras, N., Chouvarda, I., Koutkias, V., Brause, R. (eds.) ISBMDA 2006. LNCS (LNBI), vol. 4345, pp. 413–422. Springer, Heidelberg (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Sas, J., Zolnierek, A. (2009). Application of Rough Sets in Combined Handwritten Words Classifier. In: Kurzynski, M., Wozniak, M. (eds) Computer Recognition Systems 3. Advances in Intelligent and Soft Computing, vol 57. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-93905-4_60
Download citation
DOI: https://doi.org/10.1007/978-3-540-93905-4_60
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-93904-7
Online ISBN: 978-3-540-93905-4
eBook Packages: EngineeringEngineering (R0)
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.