Evaluations of Thinning Algorithms for Preprocessing of Handwritten Characters

Abstract

Thinning algorithms have played an important role in preprocessing phase which decides the success of recognition in the OCR system. This paper report on the performance of 11 thinning algorithms from the perspective of character recognition where different aspects of the performance of each algorithm like computing time, deviation from perfect 8-connectedness, and number of possible noise spurs present in the skeletons are considered.