POLYGONAL-APPROXIMATION USING A COMPETITIVE HOPFIELD NEURAL-NETWORK

Polygonal approximation plays an important role in pattern recognition and computer vision. In this paper, a parallel method using a Competi tive Hopfield Neural Network (CHNN) is proposed for polygonal approxim ation. Based on the CHNN, the polygonal approximation is regarded as a minimization of a criterion function which is defined as the arc-to-c hord deviation between the curve and the polygon. The CHNN differs fro m the original Hopfield network in that a competitive winner-take-all mechanism is imposed. The winner-take-all mechanism adeptly precludes the necessity of determining the values for the weighting factors in t he energy function in maintaining a feasible result. The proposed meth od is compared to several existing methods by the approximation error norms L2 and L(infinity) with the result that promising approximation polygons are obtained.