The ‘Modi’ script originated as a cursive variant of the script during the 17th century CE and used to write the Marathi language spoken in the Indian state of Maharashtra. Modi script evolved over time and found in many styles of writing. There is no standardization for writing characters and numerals of ‘Modi’ script but largely written without lifting the pen. The cursive nature of the script and lack of standardization in writing style pose challenges in digital recognition of documents written using Modi script including historical ones. Being largely medieval era script, modern document recognition systems lack support for recognizing handwritten texts using ‘Modi’ script. In this paper, we have described the framework of digital recognition of characters of handwritten ‘Modi’ script using empirically determined heuristics for determining the contribution of features from hybrid feature space for recognition of the ‘Modi’ character. The hybrid feature space uses normalized chain code together with feature vector encompassing a number of holes, endpoints, and zones associated with the character. The proposed framework for digital recognition of ‘Modi’ character using empirically determined heuristics provides a naïve model for recognizing a class of Indic scripts especially based on the cursive style of writing. The average and best recognition performance for the proposed method was measured to be 91.20% and 99.10% respectively.

Chain Code, OCR, Medieval Script, ‘Modi’ Script Recognition, Empirical Heuristic-based OCR

[1] R.M.K.Sinha, “Rule based contexual post-processing for Devanagari text recognition”, Pattern Recognition, 20(5), pp. 475-485, 1987.
[2] Veena Bansal, R. M. K. Sinha, “Segmentation of touching and fused Devnagari characters”, Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP`98),371 - 376, December 21-23, New Delhi, 1998
[3] R. M. K. Sinha, H. N. Mahabala, “Machine Recognition of Devanagari Script”, IEEE International Conference on Systems, Man and Cybernetics, pp. 435-441, 1979.
[4] Bansal, V. ; Sinha, R.M.K. A Complete OCR for Printed Hindi Text in Devanagari Script, Sixth International Conference on Document Analysis and Recognition, 2001. Proceedings.
[5] A S Ramteke, G S Katkar, Recognition of Off-line Modi Script : A Structure Similarity Approach, “International Journal of ICT and Management” ISSN No. 2026-6839, February 2013
[6] D. N. Besekar, A S Ramteke, “Theoretical analysis of MODI script according to recognition point of view, some issues involved with character recognition of MODI script”, International Journal of Computer Applications, February 2013
[7] D. N. Besekar, “Recognition Of Numerals Of Modi Script Using Morphological Approach”,Shodh Samiksha Aur Mulyankan Vol.III, Issue-27, april 2011
[8] Anil K. Jain, Template-based online character recognition, Pattern Recognition, Volume 34, Issue 1, January 2001, Pages 1–14
[9] D. N. Besekar, A S Ramteke, “Chain Code Approach For Recognizing Modi Neumerals , Indian Journal Of Applied Research”, December 2011
[10] Guojun Lu. “Chain Code-Based Shape Representation and Similarity measures", Visual Information Systems, 1997,Springer Berlin Heidelberg, 1997
[11] Nor Amizam Jusoh and Jasni Mohamad Zain, Application of Freeman Chain Codes: An Alternative Recognition Technique for Malaysian Car Plates”, IJCSNS International Journal of Computer Science and Network Security, VOL.9 No.11, November 2009
[12] Sadanand A. Kulkarni et al, "Offline Handwritten MODI Character Recognition Using HU, Zernike Moments and Zoning", arXiv:1406.6140 , 2014
[13] Dos Santos, Text Line Segmentation Based on Morphology and Histogram Projection, 10th International Conference on Document Analysis and Recognition, 2009. ICDAR `09.
[14] V.Di Lecce, G.Dimauro, A.Guerriero, S.Impedovo, G.Pirlo, A.Salzo, Zoning Design For Hand-Written Numeral Recognition, Proceedings of the Seventh International Workshop on Frontiers in Handwriting Recognition, 2000
[15] Sang Hak Lee, Hyung Il Koo, Nam Ik Cho, “Image segmentation algorithms based on the machine learning of features”, Published in Pattern Recognition Letters, Volume 31 Issue 14, pp. 2325-2336,October, 2010.
[16] Hung-Ming Sun, “Multi-Linguistic Optical Font Recognition Using Stroke Templates”, Proceedings of the 18th International Conference on Pattern Recognition - Volume 02 , Volume 02, August 2006.
[17] John Cowell, Fiaz Hussain, “Two template matching approaches to Arabic, Amharic and Latin isolated characters recognition”, Machine Graphics & Vision International Journal , Volume 14 Issue 2, pp. 213-232, Jan 2005
[18] Zabih, R. Kolmogorov V, “Spatially coherent clustering using graph cuts”, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Vol 2, pp. 437-444, June 2004.
[19] Mohammed Ali Qatran G, “Template matching method for recognition musnad characters based on correlation analysis”, ACIT` Proceedings, 2011
[20] Vassilis Papavassiliou, ThemosStafylakis, VassilisKatsouros, GeorgeCarayannis, “Handwritten document image segmentation into text lines and words”, Pattern Recognition, Volume 43 Issue 1, pp. 369-377 January 2010.
[21] Tanmoy Som, Sumit Saha, “A new approach for slant angle correction and character segmentation of handwritten document”, social science research network,2008.
[22] Saha, Sumit, “Handwritten Character Recognition Using Fuzzy Membership Function”, 2008 International Journal of Emerging. Technologies in Sciences and Engineering,Vol.5, No.2, pp. 11-15, Dec 2011.
[23] Swapnil Khedekar , Vemulapati Ramanaprasad , Srirangaraj Setlur , Venugopal Govindaraju, “Text–Image Separation in Devanagari Documents”, Proceedings of the Seventh International Conference on Document Analysis and Recognition, p.1265, August 03-06, 2003
[24] Rafael Palacios, Amar Gupta, “Training neural network for reading handwritten amount of checks”, Working Paper 4365-02, May 2002.
[25] S. Basavaraj Patil, N V Subbareddy, “Neural network based system for script identification in Indian documents”, Sadhana Vol. 27, Part 1, February 2002, pp. 83–97.
[26] V. Bansal, R.M.K. Sinha, “Partitioning and Searching Dictionary for Correction of Optically-Read Devanagari Character Strings”, Proceedings of the international conference on Document Analysis and Recognition (ICDAR-99), Bangalore, India, pp. 653- 656, 1999
[27] M. Hanmandlu, Harish Kumar, K. R. Murli Mohan, “Neural Based Handwritten Character Recognition”, Proceedings of the Fifth International Conference on Document Analysis and Recognition (ICDAR), pp 241 , 1999.
[28] Thein M. Ha, Horts Bunke, Off-Line, “Handwritten Numeral Recognition by Perturbation Method”, IEEE Transactions on Pattern Analysis and Machine Intelligence archive Volume 19 , Issue 5 (May 1997) pp 535 - 539
[29] B.B. Chaudhuri, U. Pal, “An OCR system to read two Indian language scripts: Bangla and Devanagari”, Proceedings of 4th International Conference on Document Analysis and Recognition, pp. 1011-1015, 1997.