Online Japanese Character Recognition Using Trajectory-Based Normalization and Direction Feature Extraction

Abstract

This paper describes an online Japanese character recognition system using advanced techniques of pattern normalization and direction feature extraction. The nor-malization of point coordinates and the decomposition of direction elements are directly performed on online tra-jectory, and therefore, are computationally efficient. We compare one-dimensional and pseudo two-dimensional (pseudo 2D) normalization methods, as well as direction features from original pattern and from normalized pat-tern. In experiments on the TUAT HANDS databases, the pseudo 2D normalization methods yielded superior per-formance, while direction features from original pattern and from normalized pattern made little difference.