BOUNDARY-BASED CORNER DETECTION USING NEURAL NETWORKS

In this paper we present a novel boundary-based corner detection appro ach using artificial neural networks (ANNs). Two neural networks are p roposed: one for detecting corner points with high curvature, and the other for detecting tangent points and inflection points that generall y have low curvature. For a given boundary point p(i), the first ANN u ses the normalized coordinates of points on the forward arm (neighbori ng points succeeding p(i)) or on the backward arm (neighboring points preceding p(i)) of the point p(i) as the input vector. The output feat ure of the network is the angle of the forward arm (or backward arm) w ith respect to the x-axis. The boundary point with sufficiently small angle between the forward and backward arms is identified as a corner. Since the feature points of tangency and inflection have relatively l ow curvature, the signs of curvature, rather than the magnitude of cur vature, for points in the neighborhood of p(i) are used as the input v ector to the second ANN. The curvature sign at each boundary point is derived from the outputs of the first ANN. The outputs of the second A NN only respond to the sign patterns of tangent points and inflection points. By using both ANNs, all features of corners, tangent points an d inflection points can be extracted from the boundary of any arbitrar y shape. Experimental results have shown that the proposed ANNs have g ood detection and localization for objects in random orientations and with moderate scale changes. Copyright (C) 1996.