Issue No. 02 - February (1979 vol. 1)

ISSN: 0162-8828

pp: 184-193

Rajjan Shinghal , MEMBER, IEEE, Department of Computer Science, Concordia University, Montreal, P.Q., Canada.

Godfried T. Toussaint , MEMBER, IEEE, School of Computer Science, McGill University, Montreal, P.Q., Canada.

ABSTRACT

In this paper a modification of the Viterbi algorithm is formally described, and a measure of its complexity is derived. The modified algorithm uses aheuristic to limit the search through a directed graph or trellis. The effectiveness of the algorithm is investigated via exhaustive experimentation on an input of machine-printed text. The algorithm assumes language to be a Markov chain and uses transition probabilities between characters. The results empirically answer the long-standing question of what is the benefit, if any, of using transition probabilities that depend on the length of a word and their position in it.

INDEX TERMS

Text recognition, Viterbi algorithm, Dictionaries, Costs, Computer science, Error correction, Councils, Probability, Decoding, Dynamic programming,text recognition, Computational complexity, contextual information, cost, feature vector, N-gram probabilities, probability of misclassification

CITATION

Rajjan Shinghal, Godfried T. Toussaint, "Experiments in Text Recognition with the Modified Viterbi Algorithm",

*IEEE Transactions on Pattern Analysis & Machine Intelligence*, vol. 1, no. , pp. 184-193, February 1979, doi:10.1109/TPAMI.1979.4766904