Hand Written Digit Recognition by Multiple Classifier Fusion based on Decision Templates Approach

Classifier fusion may generate more accurate classification than each of the basic classifiers. Fusion is often based on fixed combination rules like the product, average etc. This paper presents decision templates as classifier fusion method for the recognition of the handwritten English and Farsi numerals (1-9). The process involves extracting a feature vector on well-known image databases. The extracted feature vector is fed to multiple classifier fusion. A set of experiments were conducted to compare decision templates (DTs) with some combination rules. Results from decision templates conclude 97.99% and 97.28% for Farsi and English handwritten digits.

Authors:

• Reza Ebrahimpour
• Samaneh Hamedi

Keywords:

• Decision templates
• multi-layer perceptron
• characteristics Loci
• principle component analysis (PCA).

References:

[1] S. Impedovo, P. Wang, and H. Bunke, editors, "Automatic Bankcheck
Processing," World Scientific, Singapore, 1997.
[2] Nadal, C. Legault, R. Suen and C.Y, "Complementary Algorithms for
Recognition of totally Unconstrained Handwritten Numerals," in Proc.
10th Int. Conf. Pattern Recognition, 1990, vol. 1, pp. 434-449.
[3] CL Liu, K Nakashima, H Sako and H. Fujisawa, "Benchmarking of stateof-
the-art techniques," Pattern Recognition, vol. 36, no 10, pp. 2271-
2285, Oct. 2003.
[4] M. Shi, Y. Fujisawa, T. Wakabayashi and F. Kimura, "Handwritten
numeral recognition using gradient and curvature of gray scale image,"
Pattern Recognition, vol. 35, no. 10, pp. 2051-2059, Oct 2002.
[5] LN. Teow and KF. Loe, "Robust vision-based features and
classification schemes for off-line handwritten digit recognition," Pattern
Recognition, vol. 35, no. 11, pp. 2355-2364, Nov. 2002.
[6] K. Cheung, D. Yeung and RT. Chin, "A Bayesian framework for
deformable pattern recognition with application to handwritten character
recognition," IEEE Trans PatternAnalMach Intell, vol. 20, no. 12, pp.
382-1388, Dec. 1998.
[7] IJ . Tsang, IR. Tsang and DV Dyck, "Handwritten character recognition
based on moment features derived from image partition," in Int. Conf.
image processing 1998, vol. 2, pp 939-942.
[8] H. Soltanzadeh and M. Rahmati, "Recognition of Persian handwritten
digits using image profiles of multiple orientations," Pattern Recognition
Lett, vol. 25, no. 14, pp. 1569-1576, Oct.2004.
[9] FN. Said, RA. Yacoub and CY Suen, "Recognition of English and
Arabic numerals using a dynamic number of hidden neurons" in Proc.
5th Int Conf. document analysis and recognition, 1999, pp 237-240
[10] J. Sadri, CY. Suen, and TD. Bui, "Application of support vector
machines for recognition of handwritten Arabic/Persian digits," in Proc.
2th Iranian Conf. machine vision and image procesing, 2003, vol. 1,pp
300-307.
[11] J. Kittler and F. Roli, "Multiple Classifier Systems," in Proc Int Workshop
on Multiple Classifieer Systems, 2000, Springer.
[12] C. Suen and L. Lam, "Multiple classifier combination methodologies for
different output level," in Proc Workshop. Multiple Classifier Systems,
2000, pp.52-66.
[13] R. Duin and D. Tax, "Experiments with classifier combination rules," in
Proc Workshop. Multiple Classifier Systems, 2000, pp.16-29.
[14] F. alimoglu and E. Alpaydin, "Combining Multiple Representation for
Pen-based Hand written Digit Recognition," in Proc. 4th Int Conf.
document analysis and recognition, 1997, vol. 2, pp 637-640.
[15] K. Tumer and J. Ghosh, "Error correlation and error reduction in
ensemble classifiers," Connect Sc, vol. 8, no. 8, pp. 385-404, 1996.
[16] Gunes, M. Menard, P. Loonis and S. Renaud, "Combination,
Cooperation and Selection of Classifiers: a state of the art," Pattern
Recognition, vol. 17, no. 8, pp. 1303-1324, 2003.
[17] LI. Kuncheva, M. Skurichina and R.P.W. Duin, "An experimental Study
on diversity for bagging and boosting with linear classifiers,"
Information Fusion, vol. 3, no. 4, pp.245-258, Dec. 2002.
[18] LI. Kuncheva, JC. Bezdek and RP.W. Duin," Decision templates for
multiple classifier fusion an experimental comparison," Pattern
Recognition, vol. 34, no. 2, pp. 299-314, 2001.
[19] J. Kittler, A. Hojjatoleslami and T. Windeatt, "Weighting Factors in
Multiple Expert Fusion," in Proc. Conf. British Machine Vision, 1997,
pp. 42-50.
[20] T. Windeatt, R. Ghaderi, "Dynamic Weighting Factors for Decision
Combining," in Proc. of IEE Int. Conf. Data Fusion, Great Malvern,
UK1998, pp. 123-130.
[21] H. khosravi and E. Kabir, "Introducing a very large dataset of
handwritten Farsi digits and a study on their varieties," Pattern
Recognition letters, vol. 28, no. 10, pp. 1133-1141, July. 2007.
[22] C. J. Merz, P. M. Murphy, UCI Repository of Machine Learning
Databases 1998, http://www.ics .uci.edu/mlearn/MLRepository.html
[23] LI. Kuncheva, RK. Kounchev, RZ. Zlatev, "Aggregation of multiple
classification decisions by fuzzy templates", in Proc. 3th. European
Congres on Intelligent Technologies and Soft Computing, Germany,
August 1995, pp. 1470-1474.
[24] ID. Trier and AK. Jain, "Feature Extraction Methods for Character
Recognition- A Survey," Pattern Recognition, vol. 29, no. 4, pp. 641-
662, April. 1996.
[25] H. Takahashi, "A Neural Net OCR using geometrical and zonal pattern
features," in Proc. 1th. Conf. Document Analysis and Recognition,
1991, pp. 821-828.
[26] L.O. Jimenez, A. Morales-Morell and A. Creus, "Classification of
Hyperdimensional Data Based on Feature and Decision Fusion
Approachs Using Projection Pursuit, Majority Voting, and Neural
Networks," IEEE Trans. on Geoscience and Remote Sensing, vol. 37,
no. 3, May 1999.
[27] Y. Li, "Reforming the theory of invariant moments for Pattern
recognition," Pattern Recognition Letters, vol. 25, no. 7, pp.723-730,
July. 1992.
[28] H.A. Glucksman, "Multicategory of Patterns Represented by High-Order
Vectors of Multilevel Measurement," IEEE Transaction Computer, vol.
C-20, no. 12, pp. 1593-1598, Dec. 1997.
[29] Sh. Shahreza, M.H. Khotanzad, A. ,"Recognition Letterpress Works
Independent of Size and Displacement with Zernike Moments and
Neural Networks", in Proc. 2th Iranian Conf of Electrical
Engineering,1994, Trbiat modares Univ, vol. 5, pp. 417-424.
[30] K. Azmi, R. Kabir and E. Badi, "Recognition Printed Letters wit Zonong
Features," Iran Computer Group, vol. 1, pp. 29-37, 2003.
[31] S.H. Nabavikahrizi, R. Ebrahimpour and E. Kabir, "Application of
Combining classifiers for Recognition of Farsi handwritten digits," in
Proc. 3th Iranian Conf. Machine vision and image processing, 2004,
vol. 1, pp 115-119.
[32] Gorban, B. Kegl, D. Wunsch and A. Zinovyev (Eds.), Principal
Manifolds for Data Visualization and Dimension Reduction, LNCSE 58,
Springer, Berlin - Heidelberg - New York, 2007. ISBN 978- 3 540-
73749- 0