Optical character recognition without segmentation
In this study, an off-line optical character recognition scheme is presented. The proposed method performs the recognition by extracting the characters from the whole word, thus, avoiding segmentation process which is the most significant source of error in the recognition process.
KeywordsControl Point Hide Markov Model Recognition Rate Character Recognition Attribute Vector
- S. Impedevo, “ Fundamentals in Handwtiting Recognition”, Proc. on NATO-ASI series, Springer Verlag. 1993.Google Scholar
- F. Yarman-Vural, A. Atici, “ A Segmentation and Feature Extraction Algorithm for Arabic Script”, SPIE, Visual Communication and Image processing, vol 2, pp725–738, Orlando, 1996.Google Scholar