AFFINE-INVARIANT B-SPLINE MOMENTS FOR CURVE MATCHING

This correspondence deals with the problem of matching and recognizing planar curves that are modeled by B-splines, independently of possibl e affine transformations to which the original curve has been subjecte d (for example, rotation, translation, scaling, orthographic, and semi perspective projections), and possible occlusion, The correspondence p resents a fast algorithm for estimating the B-spline control points th at is robust to nonuniform sampling, noise, and local deformations, Cu rve matching is achieved by using a similarity measure based on the B- spline knot points introduced by Cohen et al, in 1994. This method, ho wever, can neither handle the affine transformation between the curves nor the occlusion, Solutions to these two problems are presented thro ugh the use of a new class of weighted B-spline curve moments that are well defined for both open and closed curves. The method has been app lied to classifying affine-transformed aircraft silhouettes, and appea rs to perform well.