| Typical digit recognizers classify an unknown digit pattern by computing its distance from the cluster centers in a feature space. The KNearest Neighbor (KNN) Rule assigns the unknown pattern to the class belonging to the majority of its K neighbors. These and other traditional methods adopt a uniform rule irrespective of the "difficulty" of the unknown pattern. In this paper, we propose a method ology which uses a multiple classification scheme. The classification rules of each stage are dependent on the "difficulty" of the unknown sample. Samples "far" from the center which tend to fall on the boundaries of classes are errorprone and hence "difficult". An "overlapping zone" is defined in the feature space to identify such difficult samples. We have tested this methodology on a large set (30,398) of handwritten digit images. The method described in this paper has improved the performance of the GSC digit recognizer 7 . Our method successfully reduces its error rate from 2.85% to 1.96%, i.e by 0.89%, which is more than 30% of the initial error. We have tested our method on other available classifiers and have obtained similar results. |
In: L.R.B. Schomaker and L.G. Vuurpijl (Eds.)
Proceedings of the Seventh International Workshop on Frontiers
in Handwriting Recognition, September 11-13 2000, Amsterdam,
Nijmegen: International Unipen Foundation,
ISBN 90-76942-01-3
pp. 153-165.