OFF-LINE CURSIVE HANDWRITING RECOGNITION USING HIDDEN MARKOV-MODELS

A method for the off-line recognition of cursive handwriting based on hidden Markov models (HMMs) is described. The features used in the HMM s are based on the arcs of skeleton graphs of the words to be recogniz ed. An algorithm is applied to the skeleton graph of a word that extra cts the edges in a particular order. Given the sequence of edges extra cted from the skeleton graph, each edge is transformed into a 10-dimen sional feature vector. The features represent information about the lo cation of an edge relative to the four reference lines, its curvature and the degree of the nodes incident to the considered edge. The linea r model was adopted as basic HMM topology. Each letter of the alphabet is represented by a linear HMM. Given a dictionary of fixed size, an HMM for each dictionary word is built by sequential concatenation of t he HMMs representing the individual letters of the word. Training of t he HMMs is done by means of the Baum-Welch algorithm, while the Viterb i algorithm is used for recognition. An average correct recognition ra te of over 98% on the word level has been achieved in experiments with cooperative writers using two dictionaries of 150 words each.