Handwritten characters recognition based on SKCS-polyline and hidden Markov model (HMM)

In this paper, we present a new handwritten character recognition algorithm. The proposed algorithm is based on three main steps. In the first one the original characters are segmented using separable kernel compact support (SKCS) method. In the second step a preprocessing phases: skeleton, separation, resizing, and a polyline approximation processes are then applied to the SKCS segmented characters. In the last step a handwritten hidden Markov model (HMM) is used to recognize the characters from the Cartesian coordinates of their main strokes. Simulation results are presented showing the usefulness of this new recognition method.