Image Description Via The Multiresolution Intensity Axis Of Symmetry

John M. Gauchl, Stephen M. Pizer1,2 Departments of Computer Science I and Radiology2 University of North Carolina, Chapel Hill, North Carolina, USA A fundamental approach for providing an image description in terms of visually sensible image regions is described. It involves a) the representation of the image by a structure that captures essential image information and then b) the definition of a hierarchy of components of that structure by the order of annihilation of those components as the image is continuously simplified by lowering the scale. The information-capturing "essential structure" is chosen so that image regions are associated with each structure comj>onent during the image simplification. To guarantee image simplification, successive Gaussian blurring is chosen as the means of scale lowering. We argue that an essential structure that describes shape in both the spatial and intensity dimensions will produce an image description most likely to be useful for computer or human specification of image objects. In particular, we suggest that the intensity axis of symmetry (lAS) satisfies all desirable criteria for an essential structure. With such shape-based essential structures the approach of image description via annihilation under image simplification becomes a very attractive paradigm.