Noisy digit classification with multiple specialist

Abstract

A multi-classifier formed by specialised classifiers for noise produced by an image is shown in this work. A study has been carried out in the case of structure noisy images. Classifiers based on neighbourhood criteria are used in this work, the zoning global feature and the Euclidean distance too. The experiments have been carried out with images of typewritten digits, taken from forms of the Bank of Spain. Trying to obtain a strong database to support the experiments, we have added noise to the images of the digits. The recognition rate improves from 64.58% to 96.18%. © Springer-Verlag Berlin Heidelberg 2005.