Morphological perceptron learning

Perceptrons have been used to classify patterns into different classes. Several researchers introduced a novel class of artificial neural networks, called morphological neural networks. In this new theory, the first step in computing the next state of a neuron or in performing the next layer neural network computation involves the nonlinear operation of adding neural values and their synaptic strengths followed by forming the maximum of the results. Ritter et al. (1997) have shown that the properties of morphological neural networks differ drastically from those of traditional neural network models. In this paper, the author introduces a learning algorithm for multilayer morphological perceptrons which is capable of solving arbitrary classification problems of patterns into two classes