Junction Point Elimination based Tamil Handwritten Character Recognition: An Experimental Analysis

Abstract

Optical Character Recognition is the mechanical or electronic conversion of images of typed, handwritten or printed text into machine-encoded text for online reading and study. In this OCR technique, Tamil handwritten character recognition is one of the most needed and challengeable subjects of research. An encounter between selection and extraction of proper features eliminates the issues faced by the Tamil handwritten recognition. The character and the writers complexities are the major difficulties and challenges faced by this recognition system. Eventually, the character complexity can be overcome by any pre-extraction or selection process, wherein, structured or statistical pre-extraction algorithm can be designed, discovering the essential features, in the case of writer difficulties. The challenges of Tamil handwritten characters are structure over looping; unnecessary character portion; structure discontinuations and so on. The character portions in this research are chosen by implementing a new algorithm called the Junction Point Elimination (JPE) after conscientious analysis done on the existing algorithms like Zoning, Zoning and Junction Point (ZJP) and the Junction Point (JP) based feature pre-extraction process. The novelty behind this work is introducing an algorithm in order to truncate the problem in the existing feature selection and pre-extraction algorithms. Suitable feature extraction and classification algorithm are chosen and applied on the features extracted by JPE to test whether the same can be reached in successful manner. The final analysis and experiments shows that the JPE is better than the others.

Keywords

Junction point elimination / chain code / zoning / prism tree / quad tree / support vector machine

References

[1] Boukharouba A, Bennia A. Novel feature extraction technique forthe recognition of hand-written digits. Saudi Computer Society, Applied Computing and Informatics, 2017, 13: 1926

[2] Abirami S, Manjula D. Feature string based intelligent information retrieval from Tamil document images. International Journal of Computer Applications in Technology, 2009, 35: 150-164.

[3] Antony Robert Raj M, Abirami S. A Survey onTamil handwritten character recognition using OCR techniques. The Second International Conference on Computer Science Engineering and Applications (CCSEA), 2012, 05: 115-127.

[4] Antony Robert Raj M, Abirami S. Analysisofstatisti-cal feature extraction approaches used Tamil handwritten OCR. 12th Tamil Internet Conference-INFITT, 2013 114-150.

[5] Antony Robert Raj M, Abirami S. Hybrid features based offline Tamil handwritten character recognition. 14th International Tamil Internet Conference, 2014 360-370.

[6] Antony Robert Raj M, Abirami S. Offline Tamil handwritten character recognition using chain code and zone based features. 13th Tamil Internet Conference- IN-FITT, 2014 28-34.

[7] Antony Robert Raj M, Abirami S. Offline Tamil handwritten character recognition using statistical features. Advances in Natural and Applied Sciences, 2015, 9(6): 367-374.

[8] Antony Robert Raj M, Abirami S. Strip tree based offline Tamil handwritten character recognition. Springer SIST, International Conference on ICT for Intelligent Systems, 2015 367-374.

[9] Antony Robert Raj M, Abirami S. Offline Tamil handwritten character recognition using statistical based quad tree. Australian Journal of Basic and Applied Sciences, 2016, 10(2): 103-109.

[10] Antony Robert Raj M, Abirami S, Murugappan S, Baskaran R. Prism tree shape representation based recognition of offline Tamil handwritten characters. Proceedings of the First International Conference on Computational Intelligence and Informatics (ICCII), 2016, 507: 457-470.

[11] Banumathi P, Nasira G M. Handwritten Tamil character recognition using artificial neural networks. International Conference on Process Automation, Control and Computing (PACC), 2011 1-5.

[12] Bhattacharya U, Ghosh S K, Parui S K. A two stage recognition scheme for handwritten Tamil characters. Ninth International Conference on Document Analysis and Recognition, 2007, 1: 511-515.

[13] Bremananth R, Prakash A. Tamil numerals identification. International Conference on Advances in Recent Technologies in Communication and Computing, 2009, 4: 620-622.

[14] Liu C, Suen C Y. A new benchmark on the recognition of handwritten Bangla Farsi numeral characters. Pattern Recognition, 2009, 42: 3287-3295.

[15] De Stefano C, Fontanella F, Marrocco C, Scotto di Freca A. A GA-based feature selection approach with an application to handwritten character recognition. Pattern Recognition Letters, 2014, 35: 130-141.

[16] Vamvakas G, Gatos B, Perantonis S J. Handwritten character recognition through two-stage foreground sub-sampling. Pattern Recognition, 2010, 43: 2807-2816.

[17] Pirlo G, Impedovo D. Adaptive membership functions for handwritten character recognition by Voronoi-Based image zoning. IEEE Transactions on Image Processing, 2012, 21(9): 3827-3837.

[18] Karimia H, Esfahanimehr A, Mosleh M, Mohammadian jadval ghadam F, Salehpour S, Medhati O. Persian handwritten digit recognition using ensemble classifiers. Procedia Computer Science, 2015, 73: 416-425.

[19] HP Labs India Indic and writing Datasets (2014). http://lipitk.sourceforge.net}

[20] Jagadeesh Kannan R, Prabhakar R. An improved handwritten Tamil character recognition system using octal graph. International Journal of Computer Science, 2008, 4(7): 509-516.

[21] Richarz J, Vajda S, Grzeszick R, Fink G A. Semi-supervised learning for character recognition in historical archive documents. Pattern Recognition, 2014, 47: 1011-1020.

[22] Sadri J, Yeganehzad M R, Saghi J. A novel comprehensive database for offline Persian handwriting recognition. Pattern Recognition, 2016, 60: 378-393.

[23] Dash K S, Puhan N B, Panda G. Handwritten numeral recognition using non-redundant Stockwell transform and bio-inspired optimal zoning. IET Image Processing, 2015, 9(10): 874-882.

[24] Kulkarni Sadanand A, Borde Prashant L, Manza Ramesh R, Yannawar Pravin L. Offline MODI character recognition using complex moments. Elsevier, Procedia Computer Science, 2015, 58: 516-523.

[25] Hafemann L G, Sabourin R, Oliveira L S. Learning features for offline handwritten signature verification using deep convolutional neural networks. Pattern Recognition, 2017, 70: 163-176.

[26] Das N, Reddy J M, Sarkar R, Basu S, Kundu M, Nasipuri M, Basu D K. A statistical-topological feature combination for recognition of handwritten numerals. Soft Computing, 2012, 12: 2486-2495.

[27] Das N, Sarkar R, Basu S, Saha P K, Kundu M, Nasipuri M. Handwritten Bangla character recognition using a soft computing paradigm embedded in two pass approach. Pattern Recognition, 2015, 48: 2054-2071.

[28] Pal U, Wakabayashi T, Kimura F. Handwritten numeral recognition of six popular scripts. Ninth International Conference on Document Analysis and Recognition ICDAR, 2007, 2: 749-753.

[29] Paulpandian T, Ganapathy V. Translation and scale invariant recognition of handwritten Tamil characters using hierarchical neural networks. Circuits and Systems, IEEE International Symposium, 1993, 4: 2439-2441.

[30] Pirlo G, Impedovo D. Fuzzy-zoning-based classification for handwritten characters. IEEE transactions on fuzzy systems. Pattern Recognition, 2011, 19(4): 780-785.

[31] Rajashekararadhya S V, Vanaja Ranjan P. Neural network based handwritten numeral recognition of Kan-nada and Telugu script. IEEE TENCON Conference, 2008 1-5.

[32] Rajashekararadhya S V, Vanaja Ranjan P. Efficient zone based feature extraction algorithm for handwritten numeral recognition of four popular south Indian scripts. International Journal of Theoretical and Applied Information Technology, 2008 1171-1181.

[33] Rajashekararadhya S V, Vanaja Ranjan P, Manhunath Arad-hya V N. Isolated handwritten Kannada and Tamil numeral recognition: A novel approach. First IEEE International Conference on Emerging Trends in Engineering and Technology, 2008 1192-1195.

[34] Rajashekararadhya S V, Vanaja Ranjan P. Zone-Based hybrid feature extraction algorithm for handwritten numeral recognition of two popular Indian script. World Congress on Nature and Biologically Inspired Computing, 2009 526-530.

[35] Sarkhel R, Das N, Saha A K, Nasipuri M. A multi-objective approach towards cost effective isolated handwritten Bangla character and digit recognition. Pattern Recognition, 2016, 58: 172-189.

[36] Sarkhel R, Das N, Das A, Kundu M, Nasipuri M. A multi-scale deep quad tree based feature extraction method for the recognition of isolated handwritten characters of popular indic scripts. Pattern Recognition, 2017, 71: 78-93.

[37] Mahmouda S A, Ahmad I, Al-Khatib W G, Alshayeb M, Parvez M T, Märgner V, Fink G A. Handwritten character recognition through two-stage foreground sub-sampling. Elsevier, Pattern Recognition, 2014, 47: 1096-1112.

[38] Shanthi N, Duraiswami K. A Novel SVM-based handwritten Tamil character recognition system. Journal of Computer Science, 2010, 3(9): 760-764.

[39] Shanthi N, Duraiswami K. Performance comparison of different image size for recognizing unconstrained handwritten Tamil character using SVM. Springer, Pattern Analysis and Application, 2007, 1321: 173-180.

[40] Shanthi N, Duraiswami K. A Novel SVM-based Handwritten Tamil character recognition system. Pattern Analysis Applications, 2010, 13(2): 173-180.

[41] Shyni SM, Antony Robert Raj M, Abirami S. Offline Tamil handwritten character recognition using sub line direction and bounding box techniques. Indian Journal of Science and Technology, 2015, 8(7): 110-116.

[42] Sigappi A N, Palanivel S, Ramalingam V. Handwritten document retrieval system for Tamil language. International Journal of Computer Application, 2013, 31(4): 42-47.

[43] Basu S, Das N, Sarkar R, Kundu M, Nasipuri M, Basu D K. Handwritten character recognition through two-stage foreground sub-sampling. Pattern Recognition, 2010, 43: 3507-3521.

[44] Suresh Kumar C, Ravichandran T. Handwritten Tamil character recognition using RCS algorithms. Journal of Computer Applications, 2010, 8(8): 21-25.

[45] Xu X, Wu X, Jianping W, Chenghui Z, Yuping Q. Recognition research of offline-handwritten Chinese character Based on biomimetic pattern. Procedia Engineering, 2011, 15: 5116-5120.