A New Text-Independent Handwriting Identification Method

Abstract:

For the disadvantage of traditional direction index histogram (DIH) handwriting identification method, an improved text-independent handwriting identification algorithm is presented. The handwriting image which is prepared to test needs pre-processing, then the normalized image can be obtained. Based on the features distance two factors can be extracted: the writing influence factor and the character influence factor. Compared the features of the handwriting image which is prepared to test with the features of the sample handwriting image, The better handwriting identification accuracy rate can be obtained, which provides problem-solving method for text-independent handwriting identification.

• Periodical: Applied Mechanics and Materials (Volumes 556-562)
• Pages: 4357-4360
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.556-562.4357
• Online since: May 2014
• Authors: Hong Chang Ke*, De Gang Kong
• Keywords: Handwriting Identification, Normalized Image, Text-Independent

References

[1] Y. P. Huang, S. W. Luo. Journal of Chinese In formation Processing, 2003, 17(4) : 52- 58.

[2] R. Plamondon. International Journal of Pattern Recognition And Artificial Intelligence. 2004, 8(3): 795-811.

[3] D. Yi, Q.H. Chen. Journal of computer applications. 2006, 26(l): 172-174.

[4] A. Bensefia, T. Paquet, L. Heutte. Electronic Letters on Computer Vision and Image analysis. 2005, 5(2): 72-86.

DOI: 10.5565/rev/elcvia.97

[5] P. Kammerer, M. Lettner, E. Zolda, R. Sablatnig. Pattern Recognition Letters. 2007, 28(6): 710-718.

DOI: 10.1016/j.patrec.2006.08.003

[6] Y. Zhu, T. N. Tan, Y. H. Wang. Act Automation Sinica, 2001, 27( 2 ) : 229-234.

[7] Q. Z. Ye. The signed euclidean distance transform and its applications, Proc 9th. ICPR, 1988, 495- 499.