Watson, Layne T.Shapiro, Linda G.2013-06-192013-06-191980http://hdl.handle.net/10919/20301This paper describes a new method to be used in the recognition of three-dimensional objects with curved surfaces from two-dimensional perspective views. The method requires for each three-dimensinal object a stored model consisting of a closed space curve representing some characteristic connected curved edges of the object. The input is a two-dimensional perspective projection of one of the stored models represented by an ordered sequence of points. The input is converted to a spline representation which is sampled at equal intervals to derive a curvature function. The Fourier transform of the curvature function is used to represent the shape. The actual matching is reduced to a minimization problem which is handled by the Levenberg-Marquardt algorithm [3].application/pdfenIn CopyrightTechnical reportCS80004-Rhttp://eprints.cs.vt.edu/archive/00000847/01/CS80004-R.pdf