Recognition of Off-Line Cursive Handwriting

Abstract

An entirely novel text recognition system capable of recognizing off-line handwritten Arabic cursive text having a high variability is presented [I. S. I. Abuhaiba, Ph.D. thesis, Loughborough University, 1996]. Thinned images of strokes are converted to straight-line approximations. A straight-line approximation of an off-line stroke is converted to a one-dimensional representation by a novel algorithm which aims to recover the original sequence of writing. Tokens are extracted from a one-dimensional representation of a stroke. Fuzzy sequential machines are defined to work as recognizers of tokens. The tokens of a stroke are recombined to meaningful strings of tokens. The best sets of basic shapes which represent the best sets of token strings that constitute unknown strokes are found. A method is developed to extract lines from pages of handwritten text, arrange main strokes of extracted lines in the same order as they were written, and present secondary strokes to main strokes. Presented secondary strokes are combined with basic shapes to obtain the final characters by formulating and solving assignment problems for this purpose. Some secondary strokes which remain unassigned are individually manipulated. The system was tested against the handwritings of 20 subjects yielding overall subword and character recognition rates of 55.4 and 51.1%, respectively.