Implementing Fuzziness in the Pattern Recognition Process for Improving the Classification of the Patterns Being Recognised
Correctly classifying and recognizing objects are essentially a knowledge based process. As the unpredictability of the objects to be identified increases, the process becomes increasingly difficult, even if the objects come from a small set. This variability has been taken into account by devising a fuzzy logic based approach using threshold value feature. In this paper, two methods of encoding knowledge in a system are covered-neural network and fuzzy logic-as they are currently applied to offline hand written character recognition, which is subject to high degrees of unpredictability. This paper proposes a recognition system that classifies a class of recognised patterns i.e. “partially recognised” applying fuzziness in the obtained patterns after training with backpropagation neural network and checks for the validation of the concept being proposed.
Index TermsBackpropagation network Artificial Neural Network fuzzy logic Knowledge Base
Unable to display preview. Download preview PDF.
- 1.Pandey, M.K., Dasila, N.S.: Information retrieval using artificial intelligence and fuzzy logic for hand written documents through optical character recognition (ocr). Journal of Information and Operations Management 3(1), 976–7762 (2012) ISSN: 0976–7754 & E-ISSN: 0976–7762Google Scholar
- 2.Chaudhary, B.B., Pal, U.: OCR Error Detection and Correction of an Inflectional Indian Language Script. In: Chaudhary, B.B., Pal, U. (eds.) IEEE Proceeding of 13th International Conference on Pattern Recognition 1996, August 25-29, vol. 3, pp. 245–249 (1996)Google Scholar
- 3.Mani, N., Srinivasan, B.: Application of Artificial Network Model for Optical Character Recognition. In: 1997 IEEE International Conference on System, Man and Cybernetics, Computational Cybernetics and Simulation, October 12-15, pp. 2517–2520 (1997)Google Scholar
- 4.Bansal, V., Sinha, R.M.K.: Partitioning and Searching Dictionary for Correction of Optically Read Devnagari Character Strings. In: Proceedings of the Fifth International Conference on Document Analysis and Recognition, ICDAR 1999, September 20-22, pp. 653–656 (1999)Google Scholar
- 7.Garain, U., Chaudhary, B.: IEEE Transaction on System, Man and Cybernetics- Part C: Applications and Reviews, 32 (2002)Google Scholar
- 9.Singh, R., Yadav, C.S., Verma, P., Yadav, V.: Optical Character Recognition (OCR) for Printed Devnagari Script Using Artificial Neural Network. International Journal of Computer Science & Communication 1(1), 91–95 (2010)Google Scholar
- 10.Kim, P.: Improving Handwritten Numeral Recognition Using Fuzzy Logic. IEEE Speech and Image Technologies for Computing and Telecommunications Proc. (1997)Google Scholar
- 11.Fang, X., Alouani, A.T.: Unconstrained Handwritten Numeral recognition using fuzzy Rule-Based Classifier. In: Proc. of the IEEE (2002)Google Scholar
- 12.Chan, S.-C., Nah, F.-H.: Fuzzy Neural Logic Network and Its Learning Algorithms. In: Proceedings of the 24th Annual Hawaii International Conference on System Sciences: Neural Networks and Related Emerging Technologies, Kailua-Kona, Hawaii, vol. 1, pp. 476–485 (1991)Google Scholar
- 13.Bulsari, A., Saxena, H.: Fuzzy Logic Inferencing Using a Specially Designed Neural Network Architecture. In: Proceedings of the International Symposium on Artificial Intelligence Applications and Neural Networks, Zurich, Switzerland, pp. 57–60 (1991)Google Scholar