A Gabor Wavelet Based Approach for Off-Line Recognition of ODIA Handwritten Numerals

 
 
 
  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract


    Optical Character Recognition is one of the most interesting and highly motivated areas of research, which has been very much ap-preciated in different aspect to the area of digitations world. Here in this paper we have suggested a probabilistic approach for develop-ing recognition system for handwritten Odia numerals. To report a good  level of recognition of Odia scripts is quite challenging with respect to other Indian scripts .All the procedure are sequentially enclosed to develop an recognition model and report a successful recognition accuracy. Here we have performed the analysis over to standard handwritten numeral database named as IITBBS Odia Numeral Database, which is collected from IIT Bhubaneswar. In the suggestive recognition system we have adopted a 2D-Gabor wavelet transformation approach for selection of feature vector. Apart from it we have also noted down the dimensional reduction to the obtained feature vector by sustaining to PCA. In order to predict high recognition rate we have followed up by RBF Neural Network classifier. In addition to it we have also evaluate different version of RBF like Gaussian and Polynomial. Performing over 400 samples each of 10 categories (400*10) number of Odia numeral images, we have maintained a well-defined training and testing ratio in the clas-sifier and achieved 98.02%, 96.8%.recognition rate for the reported classifiers.

     

     


  • Keywords


    OCR, Gabor Wavelet, Principal Component Analysis (PCA), Radi-al Basis Function (RBF), Neural Network.

  • References


      [1] J. Mantas “An overview of character recognition methodologies,” Pattern Recognition, vol. 19, pp. 425-430, 1986.

      [2] U. Pal, and B. B. Chaudhuri. "Indian script character recognition: a survey," Pattern Recognition, vol. 37, pp. 1887-1889, 2004.

      [3] R. Plamondon, and S.N. Srihari,“On-Line and off-line handwritten recognition: A comprehensive survey,” IEEE Trans on PAMI, vol. 22, 2000,pp. 62-84.

      [4] U. Pal, R. Jayadevan and N. Sharma, “Handwriting Recognition in Indian Regional Scripts: A Survey of Offline Techniques,” ACM Transactions on Asian Language Information Processing, vol. 11, pp. 1-35, 2012.

      [5] T. K. Bhowmik, et al. "An HMM based recognition scheme for handwritten Oriya numerals." Information Technology, 2006. ICIT'06. 9th International Conference on. IEEE, 2006.

      [6] M. Elzobi, A.A. Hamadi, Z.A. Aghbari, L. Dings, and A. Saeed, “Gabor Wave late Recognition Approach for Off-line Handwritten Arabic Using Explicit Segmentation,”, Image Processing and Communications Challenges 5, 245 Advances in Intelligent Systems and Computing 233, DOI: 10.1007/978-3-319-01622-1_29, 2014.

      [7] A. Sethy, P.K. Patra, and D.R Nayak, “Off-line Handwritten Odia Character Recognition Using DWT and PCA,” International Conference on Advance Computing and Intelligent Engineering, 2016.

      [8] K. Roy, T. Pal, U. Pal, and F. Kimura, “Oriya Handwritten Numeral Recognition System,” Proceedings of the 2005 Eight International Conference on Document Analysis and Recognition (ICDAR’05), 2005.

      [9] T.K Mishra, B. Majhi, and S. Panda, “A comparative analysis of image transformation for handwritten odiya numeral recognition,” International Conference on Advances in Computing, Communications and Informatics (ICACCI), 2013.

      [10] P. Pujari, and B. Majhi, “Recognition of Offline Handwritten Odia Numerals Using Support Vector Machine,” In International Conference on Computational Intelligence & Networks, 2015.

      [11] P.K. Patra, M. Nayak, S.K. Nayak, and N.K. Gabbak,“ Probabilistic Neural Network for Pattern Classification,” International Joint Conference on Neural Networks, 2002,pp. 1200–1205.

      [12] K.S. Dash, N.B. Puhan and G. Panda, “On Extraction of Features for Handwritten Odia Numeral Recognition in Transformed Domain,”,IEEE 2015.

      [13] A. Ray, S. Rajeswar and S. Chaudhry, “Text Recognition Using Deep BLSTM Networks”, IEEE,2015.

      [14] A. Sethy and P.K. Patra, “Off-line Odia Numeral Recognition Using Neural Network: A Comparative analysis,” IEEE International Conference on Computing Communication and Automation, ISBN: 978-1-5090-1666-2/16/$31.00

      [15] K.S. Dash, N.B. Puhan, and G. Panda, “Synthetic Handwritten Odia Numeral Database: From Shallow Hundreds to Comprehensive Thousands,” 2015 Fifth National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, 2015.

      [16] J.P. Jones, and L.A. Palmer, "An evaluation of the two-dimensional gabor filter model of simple receptive fields in cat striate cortex," J. Neurophysiol. vol. 58, pp. 1233-1258, 1987.

      [17] K. Soni, U. Kumar, S.K. Gupta, and S.L. Agrwal, “A New Gabor Wavelet Transform Feature Extraction Technique for Ear Biometric Recognition,” 2014 6th IEEE Power India International Conference (PIICON)

      [18] D.K. Vishwakarma, Prachi Rawat, and Rajiv Kapoor, “Human Activity Recognition Using Gabor Wavelet Transform and Ridgelet Transform,” International Conference on Recent Trends in Computing, vol. 57, pp. 630-636, 2015.

      [19] D. Singh, J.P. Saini, and D.S. Chauhan, “Hindi Character Recognition Using RBF Neural Network and Directional Group Feature Extraction Technique”, International Conference on Cognitive Computing and Information Processing (CCIP), 2015.

      [20] J. Hertz, A. Krogh, and R. Palmer, “Introduction to the theory of neural computation," Addison-Wesley Publishing Company, USA, 1991.


 

View

Download

Article ID: 15578
 
DOI: 10.14419/ijet.v7i2.32.15578




Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.