On-line character recognition using histograms of features and an associative memory

The paper investigates a new representation of shape and its use in handwritten on-line character recognition. This representation is based on the empirical distribution of features such as tangents, and tangent differences at distant points along the character signal. Recognition is carried out by a Kohonen associative memory (also called Kohonen self organizing feature map), trained using this representation, and the Hellinger distance, which measures the distance between distributions. We report on extensive experiments that show the pertinence of the representation and the superior performance of the scheme.