Bulanov

pp. 215-242

This article describes a workstation designed to create programs of pretreatment images for subsequent detection and recognition. It considers the methods and algorithms for solving the inverse problem of image processing, which arise, for example, with aerial drones being not equipped with a gyro stabilization system of aerial cameras. These include a problem of defocus and blurring recovery in the resulting images. A developed image processing software includes: defocusing and blurring recovery (an inverse problem solution); filtration; "joining frames" for the synthesis of the frame (merging into panorama) from the fragments. The article presents the interface of software support for most of the above algorithms and the results of studies using the numerical experiment.

The paper presents a canonical method of the swarm particles in two modifications to raise this method efficiency in solving multi-extreme problems of high dimension optimization. The method efficiency and its abovementioned modifications have been investigated in solving the test problems of optimization. One of the offered modifications has been used to solve the 28-dimensional molecular docking problem of proteases HIV1and darunaviry 3U7S as the molecules of receptor and a ligand, respectively.

Due to sharp growth of information sources from time to time a demand arises to upgrade a network of the tributaries based on synchronous digital hierarchy (SDH) thereby increasing its capacity. At the heart of an offered upgrading technique lies the method of expert evaluations, and the developed software complex serves as the tool for obtaining characteristics of a network. The complex allows us to analyse rather large number of options, being unevident even for some technical specialists owing to rather large amount of works in a network for their introduction. Approbation of the technique and complex showed their efficiency.

The paper proposes an AIS_PSO algorithm (Artificial Immune System_ Particle Swarm Optimization), which represents a modification of the hybrid algorithm based on the CLONALG immune algorithm and the algorithm of the particles of swarm (PSO). The modification is to replace the CLONALG algorithm with the same, but the modified algorithm. The AIS_PSO algorithm has been realized by MatLab software package. An efficiency of developed algorithm and software has been examined using Rosenbrock's, Himmelblau’s, and Rastrigin’s test functions. The AIS_PSO software allows a problem solution for the optimal control of vibration robot movement. The paper applies a known method to solve the problem of the optimal control of dynamic system based on reduction of this task to a task of nonlinear programming of high dimension. The study results have shown that a developed algorithm and software are high efficient.

The paper addresses the communication network constructions based on the synchronous digital hierarchy (SDH) via software complex, which uses Dijkstra algorithm taking into consideration real network conditions. These conditions include dead ends, a standby link path along the fibres and equipment, which differs from equipment with passed main link, etc. The software complex enables us to find optimal main and standby routes for link in SDH network of complicated structure taking into account these restrictions. In addition to the path routings, complex enables saving information on the network nodes, sections, and links in database (DB) i.e., actually, it represents a DB with graphic interface, thereby allowing an operator to have rapid information on its state.