Text Localisation and Handwriting Recognition

This report discusses several general aspects of text localisation and handwriting recognition. In particular, we consider their applications to extraction and recognition of numerals written on Giro forms. For text localisation, we investigate the problem of extracting text printed or written inside boxes on forms. We review a number of representative methods for solving this problem, describe the implementation of one of them, and present some experimental results obtained on real data. For handwriting recognition, we rst present the general methodology, including a new approach called perturbation method. Then we explain how the general methodology is applied to three subproblems in handwriting recognition, namely, the recognition of isolated numerals, that of numeral strings, and the recognition of cursively handwritten words drawn from a small lexicon. Experimental results show that our systems are either equivalent to or better than state-of-the-art systems. CR