Syntactic pattern classification by branch and bound search

A methodology for classifying syntactic patterns is proposed in this paper that performs a branch and bound search over a set of prototypes. The prototypes are first clustered hierarchically and the search is performed over the hierarchy. The proposed technique is applied to a pattern recognition system where images are described by the sequence of features extracted from the chain codes of their contours. A rotationally invariant string distance measure is defined that compares two feature strings. The methodology discussed in this paper is compared to a nearest neighbor classifier that uses 12,000 prototypes. The proposed technique decreases the time required to recognize a pattern by 93 percent and maintains a recognition rate of greater