Arabic words recognition with classifiers combination: An application to literal amounts

Abstract

The recognition of handwritten bank check literal amount is a problem that humans can solve easily. As a problem in automatic machine reading and interpreting, it presents a challenge and an interesting field of research. An approach for recognizing the legal amount for handwritten Arabic bank check is described in this article. The solution uses multiple information sources to recognize words. The recognition step is preformed in a parallel combination schema using holistic word structural features. The classification stage results are first normalized, and the combination schema is performed, after which using contextual information, the final decision on the candidate words can be done. Using this approach obtained results are more interesting than those obtained with individual classifiers. © Springer-Verlag Berlin Heidelberg 2004.