IJITCS Vol. 6, No. 2, 8 Jan. 2014
Cover page and Table of Contents: PDF (size: 302KB)
Full Text (PDF, 302KB), PP.58-63
Views: 0 Downloads: 0
Character Segmentation, Water Reservoir Method, Zone Segmentation, Handwritten Text Recognition
Segmentation of a word into characters is one of the important challenges in optical character recognition. This is even more challenging when we segment characters in an offline handwritten document. Touching characters make this problem more complex. In this paper, we have applied water reservoir based technique for identification and segmentation of touching characters in handwritten Gurmukhi words. Touching characters are segmented based on reservoir base area points. We could achieve 93.51% accuracy for character segmentation with this method. If the characters are neither broken nor overlapping, then this technique shall produce even better results.
Munish Kumar, M. K. Jindal, R. K. Sharma, "Segmentation of Isolated and Touching Characters in Offline Handwritten Gurmukhi Script Recognition", International Journal of Information Technology and Computer Science(IJITCS), vol.6, no.2, pp.58-63, 2014. DOI:10.5815/ijitcs.2014.02.08
[1]Casey R G, Lecolinet E. A Survey of Methods and Strategies in Character Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 18(7): 690-706.
[2]Bansal V, Sinha R M K. Segmentation of Touching and Fused Devanagari Characters. Pattern Recognition. 2002, 35(4):875-893.
[3]Jindal M K, Lehal G S, Sharma R K. Segmentation Problems and Solutions in Printed Degraded Gurmukhi Script. International Journal of Signal Processing. 2005, 2(5):258-267.
[4]Bansal V, Sinha R M K. A Devanagari OCR and A Brief Overview of OCR Research for Indian Scripts. In Proceedings of STRANS01, IIT, Kanpur, India.
[5]Garain U, Chaudhuri B B. Segmentation of Touching Characters in Printed Devanagari and Bangla scripts using Fuzzy Mulfactorial Analysis. IEEE Transactions on Systems, Man and Cybernetics-A. 2002, 32: 449- 459.
[6]Pal U, Datta S. Segmentation of Bangla Unconstrained Handwritten Text. In the proceedings of 7th International Conference on Document Analysis and Recognition, 2003, 1128-1132.
[7]Ikeda H, Ogawa Y, Koga M, Nishimura H, Sako H, Fujisawa H. A Recognition Method for Touching Japanese Handwritten Characters. In the proceedings of 5th International Conference on Document Analysis and Recognition, 1999, 641-644.
[8]Zhang D X. Extraction of Embedded and/or Line-Touching Character like Objects. Pattern Recognition. 2002, 35(11): 2453-2456.
[9]Jindal M K, Lehal G S, Sharma R K, Segmentation of Touching Characters in Upper Zone in printed Gurmukhi Script. In Proceedings of 2nd Bangalore Annual Compute Conference on 2nd Bangalore Annual Compute Conference, (Bangalore, India, January 09 - 10, 2009). COMPUTE '09. ACM, New York, NY, 1-6. DOI= http://doi.acm.org/10.1145/1517303.1517313
[10]Saba T, Sulong G, Rehman A. A survey on methods and strategies on Touched character segmentation. International Journal of Research and Reviews in Computer Science. 2010, 1(2):103-114.
[11]Reddy L P, Babu T R, Rao N V, Babu B R. Touching Syllable Segmentation using Split Profile Algorithm. 2010, 7 (3):17-26.
[12]Pal U, Belaid A and Choisy Ch. Touching numeral segmentation using water reservoir concept, Elsevier Science Inc, 2003, 24(1-3): 261-272.