LOCALIZATION OF DOMINANT POINTS FOR OBJECT RECOGNITION - A SCALE-SPACE APPROACH

The dominant points which are used for identifying an unknown object s hape are commonly taken as the positions where the maxima of the curva ture function occur on the object boundary. The peaks in the curvature function can be deduced from the corresponding zero-crossing points o f its first-order derivative. In practice, presence of noise signal in the object contour may introduce false zero-crossings in the differen tiation process, resulting in the apparent existence of false dominant points. Attenuation of the noise signal can be realized by convolving the contour function with a Gaussian filter. The width of the Gaussia n function, however, has to be properly decided to prevent unnecessary removal of the relevant dominant points. In this paper, a novel schem e for automatic determination of the filtering scale is reported. The method employs scale-space decomposition to form a basis for an explic it and quantitative measurement on the reliability of the dominant poi nt sets detected under different degree of filtering, with which the o ne exhibiting the highest score is selected. The method has been succe ssfully applied to extract the dominant point sets for different types of handtools without prior knowledge of their sizes, shapes and orien tations.