Handwritten Chinese character recognition using displacement extraction based on directional features

Abstract

This paper presents a recognition method for handwritten Chinese characters using displacement extraction based on directional features. In the proposed method, after extracting the features from an input image, the displacement is extracted by the minimization of an energy functional, which consists of the Euclidean distance and the smoothness of the extracted displacement. Experiments are performed on a set of handwritten Chinese characters (ETL-8B). The improvement of the recognition performance of the proposed method is achieved as compared with the method without the displacement extraction.