Recognition of halftone textures from the standpoint of stochastic geometry and functional analysis

A new formal approach to recognition of halftone textures based on the apparatus of stochastic geometry and functional analysis is described. A key element of it is the application of a new class of structural features of recognition, i.e., trip flight characteristics, obtained in the regime of automatic computer generation. The problem of minimization of the dimension of the feature space is considered. It holds the experimental verification of the efficiency of recognition of the halftone texture of cast iron microsections with inclusions of graphite.