Adaptive Neural Network Classifier for Extracted Invariants of Handwritten Digits

Authors

  • L. H. Keng Faculty of Computer Science and Information System Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
  • S. M. Shamsuddin Faculty of Computer Science and Information System Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia

DOI:

https://doi.org/10.32890/jict2004.3.1.1

Keywords:

Handwritten Digit, Zernike moments, adaptive activation function

Abstract

We propose an adaptive activation function of neural network classifier for isolated handwritten digits that undergo basic transformations. The utilized network is a backpropagation network with sigmoid and arctangent activation functions. The performance of network with both activation functions is compared. The results show that the network applying an adaptive activation function between layers converged much faster compared to non-adaptive activation functions with 50% iterations reduction. In this study, we also present experimental results of feature extraction between Zernike and d-geometric for better feature representations. Results show that Zernike features are better at representing isolated handwritten digits compared to d-geometric features with an accuracy up to 87%.

 

References

Agui, T., Takahashi, H., & Nagahashi, H. (1991). Recognition of handwritten Katakana in a frame using moment invariants based on neural network. IEEE International Joint Conference on Neural Networks, 1, 18-21 Nov (pp. 659—664). Singapore: IEEE Publications. http

Altuwijri, M. M., & Bayoumi, M. A. (1994). Arabic text recognition using neural networks. Proc. of IEEE International Symposium on Circuits and Systems. Systems, 6,30 May-2 June (pp. 415-418). London UK: IEEE Publications. Journal of ICT, 3 (1), pp: 1-

Cao, J. (1995). Handwritten numeral recognition with neural networks and fuzzy fusion: Doctoral Dissertation, University of Windsor.

Chung, Y. Y., Wong, M. T., & Bennamoun, M. (1998). Handwritten character recognition by contour sequence moments and neural network. Proc. of IEEE International Conference on System, Man and Cybernatics, 5, 11-14 Oct. 1998 (pp. 4184—4188). San Diego, CA USA: IEEE Publications. = Dehghan M., & Faez, K.(1997). Farsi handwritten character recognition with > moment invariants. Proceeding of 13™ IEEE International Conference on - Digital Signal Processing DSP 97, 2, 2-4 July 1997 (pp. 507-510). Santorini Greece: IEEE Publications. ~)

Hu, M. K. (1962). Visual pattern recognition by moment invariants. [RE DO Transaction on Information Theory, 8(2), 179 -187. ® Khotanzad, A. & Lu, J. H. (1990). Classification of invariant image representations using a neural network. IEEE Transactions on Acoustics, c Speech, and Signal Processing, 38(6), 1028-1038. ~)

Lam, L., & Suen, C. (1988). Structural classification and relaxation matching of > totally unconstrained handwritten zip-code numbers. Pattern Recognition, 21(1), 9-31. © Master. T. (1993). Practical neural networks recipes in C++. San Diego CA: Academic Press, Inc. = ~ Mitchel, B. T., & Gillies, A. M. (1989). A model-based computer vision system for recognizing keadweinen ZIP codes. Machine Vision and Applications, 2, O°. 231-243. —_ “+— = Moret, B. ME. (1982). Decision trees and diagrams. ACM Computer Surveys, 14(4), 593-623.

Mukundan, R., & Ramakrishnan, K. R. (1998). Moment functions in image analysis: Theory and applications. Singapore: World Scientific Publishing Co.Pte.Ltd.

Suen, C. Y., Nadal, C., Legault, R., Tuan, A. M., & Lam, L. (1992). Computer recognition of unconstrainted handwritten numerals. Proceeding of the IEEE, 80(7), 1162-1180. Journal of ICT, 3 (1), pp: 1-

Teague, M. R. (1980). Image analysis via the general theory of moments. Journal of the Optical Society of America, 70(8), 920-930. Technion-Israel Institute of Technology. (03 January 2003). Machine Learning-EE046195. Retrieved April 03, 03, from http://www. ee.technion.ac.il/courses/046195.

Ye, X.Y. (2001). A high accuracy offline handwritten numeral recognition system. Master Thesis, University of Windsor.

Zakaria, M. F. (1987). Fast algorithm for the computation of moment invariants. Partern Recognition, 20(6), 639-643. //ict.-uum.edu.my/ http

Downloads

Published

25-05-2004

Issue

Vol. 3 No. 1 (2004): Journal of Information and Communication Technology (JICT) Vol. 3, No.1 June 2004

Section

Articles

How to Cite

Keng, L. H., & Shamsuddin, S. M. (2004). Adaptive Neural Network Classifier for Extracted Invariants of Handwritten Digits. Journal of Information and Communication Technology, 3(1), 1-17. https://doi.org/10.32890/jict2004.3.1.1

Research impact

Harvested 2026-09-06
0 citations recorded so far

Counts differ between services because each indexes a different body of literature. None of them is the whole picture.

Identifiers DOI 10.32890/jict2004.3.1.1 OpenAlex W1562763756