Interactively matching hand-drawings using induction

In this paper we consider the task of matching patterns, as occur in hand-d rawn symbols and schematic diagrams, by their parts and relationships. Of p articular interest for computer vision is the integration of two approaches to the recognition by parts problem-graph matching and syntactic rule-base d approaches. A new procedure is developed, named CLARET, which matches par ts and relationships by tightly coupling the processes of matching and rule generation at run time. We have developed an interactive system for interp reting hand-drawn symbols and schematic drawings. The system operates invar iant to rotation, scale, and position and projects images onto a drawing ca nvas. The procedure is analyzed for its ability to accommodate new symbols and answer orientation queries, and it is compared empirically with machine learning techniques. (C) 1999 Academic Press.