Neural Network Clustering Technology for Cartographic Images Recognition

Abstract

It is offered the information system of recognition of cartographic images of soil massifs and classification of landscape areas by types of soil massifs using the neural network. Here was described approaches to architecture design, teaching methods, data preparation for teaching, training and neural network testing. The functional scheme of the neural network is developed, which consists of the input, hidden and output layer, collecting and processing of data, and training algorithm. The analysis of efficiency, speed and accuracy of work of a neural network as a part of information technology is carried out.