Fast thinning algorithm based on improved SOG (SOG)

Abstract

We propose Improved Self-Organized Graph(Improved SOG: SOG " ) thinning method, which maintains excellent thinning results of Self-Organized Graph(SOG) built from Self-Organizing features Map and improves the pl'rformance of the existing SOG by using II new incremental learning method of Kohonen features map. In the experiments, this method shows thl' thinning results to be equal (0 those of SOG but the time compleKity O«(logM») to be superior to that of SOG. Th .. refore, the proposed method is useful for tbe feature extraction from digils and characters in the preprocessing step.