FEATURE-EXTRACTION IN SHAPE-RECOGNITION USING SEGMENTATION OF THE BOUNDARY CURVE

We present a new method for feature extraction of two-dimensional shap e information based on segmentation of the boundary curve. This approa ch partitions closed shapes into segments and finds their angular span s. The number of segments and the angular spans form the first two fea ture parameters of a given shape. Fourier coefficients of all segments constitute the final feature parameters. The algorithm renders the sh apes independent of scale, rotation and translation, The main advantag e of this method is to speed up substantially the recognition process of the shapes, mainly because it is possible to design the classificat ion rule in a hierarchical way. It is therefore suitable for objects t o be sorted in a factory environment where the silhouette boundary sup plies sufficient information for identification. (C) 1997 Elsevier Sci ence B.V.