SHAPE REPRESENTATION AND RECOGNITION FROM MULTISCALE CURVATURE

We present a technique for shape representation and the recognition of objects based on multiscale curvature information, It provides a sing le framework for both the decomposition and recognition of both planar curves as well as surfaces in three-dimensional space, The decomposit ion operation simultaneously performs data interpolation, data smoothi ng, and segmentation. The unification of these three stages results in a smoothing operation that is coupled with the primitives to be used in description. Each of the minimization operators, in addition to hav ing a curvature tuning, also has a different spatial sensitivity funct ion, As a result, the different possible descriptions capture informat ion at multiple spatial scales, This allows a single region of an obje ct to be described in more than one way, when appropriate. The practic ality of the ensuing representation is demonstrated by the recognition of planar curves. A matching strategy based on dynamic programming is used, The results illustrate the manner in which a continuous spectru m of similar objects can be defined, ranging from those that are very similar to a target to those that are very different from it. (C) 1997 Academic Press.