The increasing of image classification efficiency is proposed through the introduction of effective ways of bits description weights vector for etalon images, which is presented as a set of key point descriptors. Mathematical models of logical processing are proposed in order to determine weight coefficients and construct similarity measures for image descriptions. Software modeling of processing methods is performed; classification efficiency using modified approaches in applied image databases is evaluated. Results of modeling have been confirmed the increase in classification performance for the proposed modifications to logical data processing.
No takes yet. Share an insight, caveat, or question.
Gorokhovatskyi et al. (2020) studied this question.