Cellular neural networks with opposite-sign templates for image thinning

This paper presents image thinning algorithms using cellular neural network s (CNNs) with one- or two-dimensional opposite-sign templates (OSTs) as wel l as non-unity gain output functions. Two four-layer CNN systems with one-d imensional (l-D) OSTs are proposed for image thinning with 4- or 8-connecti vity, respectively. A CNN system, which consists of an eight-layer CNN with two-dimensional (2-D) OSTs followed by another four-layer CNN with 2-D OST s, is constructed for image thinning with 8-connectivity, in which designs of B- and I-templates are simpler than in CNNs with l-D OSTs. In the aforem entioned designs, parameter values of l-D OSTs are chosen to make CNNs oper ate with thinning-like property 1 (TL-1), and those of 2-D OSTs with 2-D th inning-like property (2-DTL). Simulation studies show that these CNN system s have a good image thinning performance. Copyright (C) 1999 John Wiley & S ons, Ltd.