Handwritten and Printed Devanagari Compound using Multiclass SVM Classifier with Orthogonal moment Feature

Devanagari character plays a vital role in the research area. Number of technique has been adopted in previous decades and still some new are arising to get good results from recognition system. In Devanagari, Compound character are complex in structure, they are written by combination two or more character. Due to complex structure in it, it gives a challenging task to the researchers. The occurrence of compound in the script is upto 8 to 10%. In this research paper, a recognition system for handwritten and printed Devanagari Compound Character is proposed bases on orthogonal moment i.e. Legendre and Zernike moment as a feature are used to recognize the Devanagari character. The character image is preprocessed and normalized to 30X30 pixel sizes, and used for structural classification where the character is classified into three classes. After classifying the character it is partitioned and from each zone moment feature are extracted. The proposed system is trained and tested on 27000 handwritten collected from writer of different profession, 10800 printed Devanagari basic and compound character database under APS Designer Software. For classification fivefold cross validation test is used with SVM to obtain average percentage of recognition accuracy. The RR for Printed is 98.42 % (basic) and 98.31% (compound) and for Handwritten is 98.51% (Basic) and 98.30% (Compound) by SVM. The total number of classes is 48 Basic character and 60 Compound + split character. The obtained results were compared with some related existing approaches. Owing to the proposed technique, the results obtained show higher efficiency regarding classifier accuracy.

[1]  Vanita Mane,et al.  Handwritten character recognition using elastic matching and PCA , 2009, ICAC3 '09.

[2]  Fumitaka Kimura,et al.  Indian Multi-Script Full Pin-code String Recognition for Postal Automation , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[3]  Venu Govindaraju,et al.  Half-Against-Half Multi-class Support Vector Machines , 2005, Multiple Classifier Systems.

[4]  Tetsushi Wakabayashi,et al.  Comparative Study of Devnagari Handwritten Character Recognition Using Different Feature and Classifiers , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[5]  Prabin Kumar Bora,et al.  A Comparative Study on Discrete Orthonormal Chebyshev Moments and Legendre Moments for Representation of Printed Characters , 2004, ICVGIP.

[6]  Shaila Apte,et al.  Multistage Handwritten Marathi Compound Character Recognition Using Neural Networks , 2011 .

[7]  Veena Bansal,et al.  Partitioning and searching dictionary for correction of optically read Devanagari character strings , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[8]  B. Fei,et al.  Binary tree of SVM: a new fast multiclass training and classification algorithm , 2006, IEEE Transactions on Neural Networks.

[9]  Bidyut Baran Chaudhuri,et al.  A complete printed Bangla OCR system , 1998, Pattern Recognit..

[10]  Mahantapas Kundu,et al.  Combining Multiple Feature Extraction Techniques for Handwritten Devnagari Character Recognition , 2008, 2008 IEEE Region 10 and the Third international Conference on Industrial and Information Systems.

[11]  Latesh G. Malik,et al.  Fine Classification & Recognition of Hand Written Devnagari Characters with Regular Expressions & Minimum Edit Distance Method , 2008, J. Comput..

[12]  Ujjwal Bhattacharya,et al.  A Two Stage Recognition Scheme for Handwritten Tamil Characters , 2007 .

[13]  Mita Nasipuri,et al.  A Two Stage Classification Approach for Handwritten Devnagari Characters , 2010, International Conference on Computational Intelligence and Multimedia Applications (ICCIMA 2007).

[14]  Ishwar K. Sethi,et al.  Machine recognition of constrained hand printed devanagari , 1977, Pattern Recognit..

[15]  Richard J. Prokop,et al.  A survey of moment-based techniques for unoccluded object representation and recognition , 1992, CVGIP Graph. Model. Image Process..

[16]  Arun K. Pujari,et al.  An intelligent character recognizer for Telugu scripts using multiresolution analysis and associative memory , 2004, Image Vis. Comput..

[17]  Tetsushi Wakabayashi,et al.  Off-Line Handwritten Character Recognition of Devnagari Script , 2007 .

[18]  Hamid Reza Boveiri.  Persian Printed Numeral Characters Recognition Using Geometrical Central Moments and Fuzzy Min-Max Neural Network , 2010 .

[19]  Fumitaka Kimura,et al.  Recognition of Off-Line Handwritten Devnagari Characters Using Quadratic Classifier , 2006, ICVGIP.

[20]  Bidyut Baran Chaudhuri,et al.  A system for Indian postal automation , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[21]  Mahantapas Kundu,et al.  Recognition of Non-Compound Handwritten Devnagari Characters using a Combination of MLP and Minimum Edit Distance , 2010, ArXiv.

[22]  Fumitaka Kimura,et al.  Oriya handwritten numeral recognition system , 2005, Eighth International Conference on Document Analysis and Recognition (ICDAR'05).

[23]  Veena Bansal,et al.  On how to describe shapes of Devanagari characters and use them for recognition , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).

[24]  Mahantapas Kundu,et al.  Study of Different Features on Handwritten Devnagari Character , 2009, 2009 Second International Conference on Emerging Trends in Engineering & Technology.

[25]  R. D. Sudhaker Samuel,et al.  A simple and efficient optical character recognition system for basic symbols in printed Kannada text , 2007 .

[26]  S. M. Murtoza Habib,et al.  A High Performance Domain Specific Ocr For Bangla Script , 2008 .

[27]  S. Shelke,et al.  A novel multistage classification and Wavelet based kernel generation for handwritten Marathi compound character recognition , 2011, 2011 International Conference on Communications and Signal Processing.

[28]  T. R. Sontakke,et al.  Rotation, scale and translation invariant handwritten Devanagari numeral character recognition using general fuzzy neural network , 2007, Pattern Recognit..

[29]  Satish Kumar,et al.  Performance Comparison of Features on Devanagari Hand-printed Dataset , 2009 .

[30]  Santanu Chaudhury,et al.  Devnagari numeral recognition by combining decision of multiple connectionist classifiers , 2002 .

[31]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[32]  Bidyut Baran Chaudhuri,et al.  Indian script character recognition: a survey , 2004, Pattern Recognit..

[33]  M. Teague.  Image analysis via the general theory of moments , 1980 .

[34]  David G. Stork,et al.  Pattern Classification , 1973 .

[35]  Veena Bansal,et al.  Integrating knowledge sources in Devanagari text recognition system , 2000, IEEE Trans. Syst. Man Cybern. Part A.

[36]  Soo-Young Lee,et al.  Support Vector Machines with Binary Tree Architecture for Multi-Class Classification , 2004 .

[37]  B. B. Chaudhuri.  Digital document processing : major directions and recent advances , 2006 .

[38]  Karbhari V. Kale,et al.  COMPARISON OF CLASSIFIERS FOR GUJARATI NUMERAL RECOGNITION , 2011 .

[39]  Tetsushi Wakabayashi,et al.  Handwritten Bangla Compound Character Recognition Using Gradient Feature , 2007 .

[40]  S.N. Nawaz,et al.  An approach to offline Arabic character recognition using neural networks , 2003, 10th IEEE International Conference on Electronics, Circuits and Systems, 2003. ICECS 2003. Proceedings of the 2003.