Experiments with a Self-Supervised Adaptive Classification Strategy in On-Line Recognition of Isolated Handwritten Latin Characters
Jorma T. Laaksonen, Jarmo Hurri, Erkki Oja, Jari A. Kangas · 1998
1 Introduction We present the results of a series of experiments which assessed the development of an adaptive classifier for on-line recognition of handwritten Latin characters. We have had a series of work hypotheses on our mind. First, that information about the typical ways characters are written can be automatically extracted- or distilled- from a database containing data from numerous 1