The subject of research of the article is the methods of image classification according to the set of descriptors of key points in computer vision systems. The aim is to increase the efficiency of classification by introducing a multicomponent data model on a set of descriptors for the base of reference images. Applied methods: ORB detector and descriptors, apparatus of set theory and vector space, metric models for determining the relevance of sets of multidimensional vectors, elements of probability theory, software modeling. Results are obtained: a modified method of image classification based on the introduction of a multicomponent model for data analysis with a system of centers is developed, methods of constructing a set of data centers are identified, the most effective is the set medoid and centers based on it. The effectiveness of the modification significantly depends on the method of forming the centers, the applied classification model, as well as on the data itself. The best results were shown by the classification with the integrated indicator separately for each of the standards in the form of the sum of the values of the distributions for the set of centers; experimentally tested the effectiveness of the classification, confirmed the efficiency of the proposed method. The practical significance of the work is the construction of classification models in the transformed data space, confirmation of the efficiency of the proposed modifications on the examples of images, the creation of software for the implementation of developed classification methods in computer vision systems.

The subject of research in the article is dynamic control systems with optimal slow motions. The goal of the work is to obtain an asymptotic approximation of the control in the form of feedback, which, not being uniform in the domain of the system definition, forms slow motions of the system uniformly close to optimal ones. The objectives of the study are to conduct an asymptotic analysis of the controller for small values of the parameters. Applied methods: methods of minimization of quadratic functions and methods of matrix algebra. The obtained results: the problem of the optimal equation with two small singularly exciting parameters is considered. Requirements for the characteristics and controllability of the selected system have been introduced. The problem under consideration, in contrast to well-known studies, is connected with a fundamental problem: as the system parameters tend to zero, certain components of the matrix that satisfies the Ricatti equation, due to the limiting condition for it, acquire singularities in certain time intervals. The practical significance of the work lies in the fact that with the use of minimization methods, general methods are obtained for constructing a uniform region of asymptotics for optimal control systems with two small singular-exciting parameters with respect to another small parameter.

The subject of the research is methods of identifying the state of the Computer System. The object of research is the process of identifying the state of a computer system for information protection. The aim of the research is to develop the method for identifying the state of a computer system for information protection. This article is devoted to the development of method (boosting ensemble) to increase the accuracy of detecting anomalies in computer systems. Methods used: artificial intelligence methods, machine learning, decision tree methods, ensemble methods. The results were obtained: a method of computer system identification based on boosting ensemble with special preprocessing procedure is developed. The effectiveness of using machine learning technology to identify the state of a computer system has been studied. Experimental researches have confirmed the effectiveness of the proposed method, which makes it possible to recommend it for practical use in order to improve the accuracy of identifying the state of the computer system. Conclusions. According to the results of the research, ensemble classifier of computer system state identification based on boosting was proposed. It was found that the use of the proposed classifier makes it possible to reduce the variance to 10%. In addition, due to the optimization of the initial data, the efficiency of identifying the state of the computer was increased. Prospects for further research may be to develop an ensemble of fuzzy decision trees based on the proposed method, optimizing their software implementation.

Today, a variety of information about forest ecosystems can be obtained using remote sensing methods. The use of space data for forest monitoring is cost-effective because it allows you to quickly obtain the objective information needed by foresters to solve practical problems. Satellite data provide wide coverage of forest lands, high accuracy of results, as well as high frequency of data obtained. Space images of the Ovruch district of the Zhytomyr region of Ukraine in the summer of 2020 were selected for the study. Determination of breed composition was carried out by the methods of controlled classification, namely the Bayesian classifier. It was found that 70 % of forests are pine, less aspen, hornbeam, birch, alder and ash tree species. According to statistics, during 2000-2020, 51.4 thousand hectares of forest plantations in Ukraine were damaged and destroyed by forest fires. Therefore, objective and timely information on the consequences of fires is needed to solve a wide range of applied problems of forestry. An important task in assessing the environmental and economic damage caused to forestry as a result of forest fires is to determine the area of damaged forests. The paper considers technologies for determining the area of the forest where the fire took place, using space images of the Landsat 8 satellite. The normalized NBR fire index before and after the fire and the DNBR index are used to identify areas burned by fire and impression levels. To predict forest fires, a mathematical model based on Bayes theorem was created and a thematic map with fire hazard classes on a quarterly basis was created. To check the accuracy of the results of the created forecast model, the thematic map was combined with a layer of defined fire areas. This software product is quite flexible and versatile, it can be easily adapted for use not only to identify burned forest lands, but also for other areas.

The main task of path search methods is to generate a possible path through a section of the environment while bypassing obstacles and minimizing the path length. Existing methods focus on the shortest geometric path to the target and do not take into account various parameters, such as energy consumption or the complexity of sections of the path. This article proposes a method for finding a path and determining its optimality using Petri nets. In addition to generating solutions based on several parameters, this method allows you to expand the visual feedback. When calculating the optimality of the model using the geometric distance from the entry point of the robot to the goal along a certain route and the cost of the abstract amount of energy to move each route. The object of research is Petri nets and their use to model the process of creating routes and finding a way for autonomous robots. The subject of research is the mathematical apparatus of Petri nets and the feasibility of their use in modeling the process of creating routes and finding a way for autonomous robots. The research aims to demonstrate the benefits of Petri nets for visual modeling of the process of finding the path and motion of autonomous robots. Conclusions. The presented technique is suitable for use in modeling the process of finding a way and provides comprehensive statistics for further processing and analysis.

Modern information technologies include the use of server systems, virtualization technologies, communication tools for distributed computing and development of software and hardware solutions of data processing and storage centers, the most effective of such complexes for managing heterogeneous computing resources are hybrid GRID- distributed computing infrastructure combines resources of different types with collective access to these resources for and sharing shared resources. The article considers a multi-agent system that provides integration of the computational management approach for a cluster Grid system of computational type, the nodes of which have a complex hybrid structure. The hybrid cluster includes computing modules that support different parallel programming technologies and differ in their computational characteristics. The novelty and practical significance of the methods and tools presented in the article are a significant increase in the functionality of the Grid cluster computing management system for the distribution and division of Grid resources at different levels of tasks, the ability to embed intelligent computing management tools in problem-oriented applications. The use of multi-agent systems for task planning in Grid systems will solve two main problems - scalability and adaptability. The methods and techniques used today do not sufficiently provide solutions to these complex problems. Thus, the scientific task of improving the effectiveness of methods and tools for managing problem-oriented distributed computing in a cluster Grid system, integrated with traditional meta-planners and local resource managers of Grid nodes, corresponding to trends in the concept of scalability and adaptability.

The subject of the article is the development of a method for implementing the arithmetic operation of adding the residuals of numbers, which are represented in the system of residual classes (RNS). This method is based on the use of positional binary adders. The purpose of the article is to improve the performance of computer systems (CS) and their components by introducing new ways of organizing calculations based on the use of RNS. Tasks: to analyze and identify the shortcomings of the existing number systems that are used in the construction of computer systems and components; explore possible ways to eliminate the identified deficiencies; explore the structure of binary positional adders, taking into account the scheme for adding two residues of numbers modulo RNS; to develop a method for constructing adders modulo RNS, which is based on the use of a set of binary single-digit positional adders. Research methods: methods of analysis and synthesis of computer systems, number theory, coding theory in RNS. The following results are obtained. The paper shows that one of the promising ways to improve the performance of the CS is the use of RNS. The mathematical basis of RNS is the Chinese remainder theorem, which states that an integer operation on one large modulus can be replaced by a set of operations on coprime small modules. This opens up broad prospects for optimizing calculations. On the one hand, it is possible to significantly simplify the performance of complex and cumbersome calculations, including on low-resource computing platforms. On the other hand, calculations for different modules can be performed in parallel, which increases the performance of the CS. Conclusions. The article considers the operation of adding two numbers. This operation is the basis for both traditional positional number systems and RNS, i.e. forms the computational basis of all existing CS components. A new method for calculating the sum of the residuals of numbers modulo an arbitrary is proposed, and examples are given that clearly demonstrate the effectiveness of the proposed method. This method can be used in various computer applications, including for improving computing performance, ensuring fault tolerance, etc.

The subject of this research is noise reduction methods in the sound sequence as a part of the proposed speech-to-text (STT) module for converting a verbal lecture or a lesson into a written text form on digital educational platforms. The goal is to investigate the influence of noise reduction methods on the operation of the acoustic signal recognition system. 3 methods of noise reduction were considered for integration in the proposed acoustic artifact recognition system and for the researching: spectral subtraction method; fast Fourier transform; Wiener filter with software modeling of every method. The obtained results: after testing the system with integrated noise reduction modules in it, based on the fast Fourier transform, Wiener filter and spectral subtraction method, it was concluded that the module using the Wiener filter improves the identification results by 25%, which is the highest result. However, performance testing has shown that fast Fourier transform is the fastest method. The practical significance of the work is – the identifying acoustic events system was developed, different noise reduction methods were integrated and researched into the module for converting a verbal lecture or a lesson into a written text form in a proposed system with the aim of increasing of speed and accuracy.

Detection and recognition of loud sounds and characteristic noises can significantly increase the level of safety and ensure timely response to various emergency situations. Audio event detection is the first step in recognizing audio signals in a continuous audio input stream. This article presents a number of problems that are associated with the development of sound event detection systems, such as the deviation for each environment and each sound category, overlapping audio events, unreliable training data, etc. Both methods for detecting monophonic impulsive audio event and polyphonic sound event detection methods which are used in the state-of-the-art sound event detection systems are presented. Such systems are presented in Detection and Classification of Acoustic Scenes and Events (DCASE) challenges and workshops, which take place every year. Beside a majority of works focusing on the improving overall performance in terms of accuracy many other aspects have also been studied. Several systems presented at DCASE 2021 task 4 were considered, and based on their analysis, there was a conclusion about possible future for sound event detection systems. Also the actual directions in the development of modern audio analytics systems are presented, including the study and use of various architectures of neural networks, the use of several data augmentation techniques, such as universal sound separation, etc.

The article considers the possibility of determining the source of higher harmonics of currents and voltages in the electrical network in the directions of power of the respective components. The research purpose is evaluating the effectiveness of theoretical methods, computer modeling of a fragment of the electrical network and electrical load receivers operating in parallel. Different combinations of active, reactive and nonlinear loading are investigated. Larionov scheme was used as a nonlinear load. According to the results of the study, conclusions were made about the effectiveness of the method used. Thus, in particular, in the case of parallel operation of nonlinear and mixed active-nonlinear load, the source of higher harmonics is determined incorrectly. Suggestions for ways to improve the algorithm are made.

The fiber-optic communication lines and fiber-optic transmission systems using fiber-optic cable, receiving and transmitting optical modules based on WDM and DWDM (Wavelength Division Multiplexing & Dense WDM) technologies are studied. A method for calculating the transmission characteristics fiber-optic communication lines is proposed and relationships are obtained that establish an analytical relationship between the length regeneration section and the transmission rate. This paper discusses the study and analysis indicators fiber-optic communication lines using spectral WDM and DWDM technologies.

Traditional technologies for reliability analysis of semi-Markov systems are limited to obtaining a stationary state probability distribution. However, when solving practical control problems in such systems, the study of transient processes is of considerable interest. This implies the subject of research - the analysis of the laws of distribution of the system states probabilities. The goal of the work is to obtain the desired distribution at any time. The complexity of the problem solving is determined by the need to obtain a result for arbitrary distribution laws of the duration of the systems stay in each state before leaving. An easy-to-implement method for the analysis of semi-Markov reliability models has been suggested. The method is based on the possibility of approximating probability-theoretic descriptions of failure and recovery flows in the system using the Erlang distribution laws of the proper order. The developed computational scheme uses the most important property of Erlang flows, which are formed as a result of sieving the simplest Poisson flow. In this case, the semi-Markov model is reduced to the Markov one, which radically simplifies the analysis of real systems.

The paper presents an intelligent method for detecting UAV spoofing. A distinctive feature of the method is the use of subtrajectory calculation technology based on visual odometry subtrajectories and GPS positions in a sliding window, taking into account the intelligent estimation of the optical flow and the formation of UAV “Ego-movement” descriptors. In the course of the study, an analysis and comparative studies of a wide range of UAV spoofing methods were carried out, the most frequently recommended and practically used methods were identified. The conclusion is made about the relevance of the problems of GPS spoofing. The analysis of methods of protection against UAV GPS spoofing has been carried out. Promising directions for intelligent detection of UAV spoofing using methods and means of visual odometry are identified. In the course of studying methods for fixing input data, an approach was proposed for estimating the optical flow using a sliding window. At the same time, the need for intelligent processing of input data is argued. The estimation of the optical flow and the formation of descriptors was carried out using recurrent convolutional neural networks. As a result, a block diagram of the UAV spoofing detection method was developed. This allowed us to study the developed method. The results of the experiment for two spoofing scenarios showed the efficiency of estimating the positions of at least two of the three indicators under the conditions of using sliding windows of size 15 or more, with a time delay of half the window size.

The experience of the International Medical Device Regulators Forum in the use of products for oxygen therapy based on the study of guidelines, international and European regulations on medical devices is considered. Emphasis is placed on issues related to the metrological support of pulse oximeters. The global situation with the pandemic increases the relevance of selected topics. Continuous monitoring of blood oxygen saturation is one of the most important elements in the diagnosis and treatment of COVID-19. Despite the fact that the use of pulse oximeters does not require special medical knowledge, calibration and special maintenance, it is still necessary to control their reliability and accuracy of measurement. The study of pulse oximeter MD300M in accordance with the requirements of DSTU 8893:2019 "Metrology. Pulse oximeters. Calibration Method" 2020. Analysis of MD300M pulse oximeter calibration results revealed that this pulse oximeter was calibrated by the saturation and pulse measurement channel, as the maximum deviation of the measured saturation value from the pulse oximeter calibration measure does not exceed the pulse oximeter measurement documentation specified in the operating documentation. The discreteness of the MD300M pulse oximeter and the discreteness of the reference device MPPO-2, which are specified in the passports for the respective devices, have the greatest influence on the measurement uncertainty.

The subject of the review is methodology of the subsystem verification and printing 3 D -model online learning system mixed type. To do this, a review of analogues, technologies, stages of printing were identified. Due to the development of technology, the educational process is being transformed. Education uses blended learning, part of which is distance learning. The object of research is use of additive technologies, which can make the learning process more motivating. Thus, if in distance education there is an opportunity to develop a 3D model online, check it for fidelity, send the model to print, it optimizes the learning process. The aim is to design with IP topics that uses the additive technologies in the educational process. Methods used: IDEF- diagram describing the function of the system; authentication rules, verification of 3 D models, sending the model to print, selecting a device online, and basic screen forms. Conclusions. The development of innovative thinking in higher education students should become a priority of modern higher education, and the introduction of new elements in modern education is inevitable. And given the development of 3D printing technologies, additive technologies are the most promising for the use of visualization in online and mixed teaching.

The subject of research in the article is the methods of testing digital devices implemented on programmable logic devices (PLDs). The aim of the research is to substantiate the approach to diagnosing microprocessor systems based on PLD, which is a further development of the self-diagnostics method. The following tasks are solved in the article: the analysis of existing methods for testing digital devices is conducted, their advantages and disadvantages are disclosed; analyzed the existing approaches to the diagnosis of microprocessor systems, which are based on integrated circuits with a programmable structure; an approach to diagnosing microprocessor systems based on PLDs is proposed, which is based on the principle of interaction between processors by introducing a service processor into a multiprocessor system. The following results have been obtained: it has been proved that the implementation of the proposed models of interaction between processors in a multiprocessor system will provide the service processor with diagnostic information about the technical state of the system. It is noted that the availability of diagnostic information about the technical condition of the microprocessor system is the basis for making a decision to restore the system by reprogramming the PLD. Conclusions: the problem of diagnosing microprocessor systems in the PLD basis is considered. The analysis of existing methods for testing integrated circuits with a programmable structure has been conducted, their advantages and disadvantages have been noted. It is proposed to use the built-in service processor as a diagnostic device, the main function of which is to collect diagnostic information and make a decision on the reconfiguration of the microprocessor system in order to quickly restore its functioning automatically.

The actuality of the power to improve the accuracy of the results was determined in order to make a decision about the process of testing the software security. An analysis of the methods of support for making a decision was carried out. The necessity and feasibility of improving the accuracy of the results was determined in case of further software security inconsistencies in the minds of the fuzziness of input and intermediate data. With this method, on the basis of the mathematical apparatus of fuzzy logic, the method of support for making a decision about the security of software security has been developed. The main feature of this method is the synthesis of an improved method of generating the initial vibration in the process of starting a piece of neural string. Within the framework of the model, the next stages of follow-up are reached. For the mathematical formalization of the process of accepting the decision and designation of the input data, the model of forming the vector in the input data was developed. Depending on this model for shaping the input data, an anonymous sign of potential inconsistencies and undeclared possibilities of the PP is valid until the data of PVS-Studio Analysis Results. To improve the accuracy of the classification of data collected, the method of creating a piece of neural array has been improved, which is modified by the method of generating a sample, which is being developed. This generation method includes three equal generations: generation of the initial vibration, generation of the initial butt and generation of a specific value of the safety characteristic. This made it possible to increase the accuracy of classification and acceptance of the solution by 1.6 times for positive elements in the selection by 1.2 times for negative elements in the selection. To confirm the effectiveness of the development of the method of support for the decision on how to ensure software security, a ROC-analysis was carried out over the course of the above procedures. The results of the experiment confirmed the hypothesis about the efficiency of the divided method of support to make a decision about the security of PZ up to 1.2 times equal to the methods, which are based on the position of discriminant and cluster analysis.

The goal of the work. Development of methods for performing basic arithmetic operations with interval complex numbers, which are presented in hyperbolic form, their modulus and argument. Results. The paper considers the method of extending interval numbers defined in hyperbolic form (hyperbolic interval numbers) to the field of complex numbers. To do this, the real and imaginary part of a complex number is presented in the form of a hyperbolic interval number. The connections between the representation of interval numbers in the classical form, the CENTER-RADIUS system and the hyperbolic form are established. Methods of performing basic arithmetic operations with hyperbolic complex numbers are proposed, namely: addition, subtraction, multiplication and division. A method of raising the positive interval number of a complex interval number defined in a hyperbolic form to an integer positive degree is proposed. Methods for calculating the modulus and argument of a complex number defined in hyperbolic form are proposed. A method for determining the root of a degree from an interval complex number represented in hyperbolic form is proposed. Using the connections between hyperbolic and trigonometric functions, a form of representation of an interval number in trigonometric form is proposed. It is established that it is most expedient to perform addition and subtraction actions with complex interval numbers, which have a classical form or are defined in the CENTER-RADIUS system. The operations of multiplication, division and elevation to an integer power are most expedient to perform with complex interval numbers which are defined in hyperbolic form. The operation of calculating the root of a degree from an interval complex number, presented in hyperbolic form, is most expedient to perform with the combined use of the representation of the interval number in the system CENTER-RADIUS and in hyperbolic form.

In the modern world, the image is becoming a more powerful means of communication. In the era of digitalization, people can express their thoughts, feelings, and more through images. The main difficulty of the author is always how to prove his authorship. Therefore, the development of the latest technologies for confirming authorship of digital images is relevant. The object of the study is the process of confirming the fact of the right to own a digital image. The subject of the study is the information technology of proof of authorship for digital images. Taking into account modern trends and results of using blockchain for copyright protection, this paper proposes an information technology for confirming ownership of digital images based on blockchain technology and digital watermarks to provide a reliable guarantee of copyright establishment. Proposed digital image ownership information technology using digital watermarking, blockchain, image hash function and IPFS to create a brand new decentralized ownership verification technology in the digital age of the Internet. As far as file types are concerned, this article only mentions digital image copyright management. In the future, the technology could be extended to include audio, video, and other types of media files to create a single system for verifying ownership of any digital content.

The object of research is the process of improving the network architecture of enterprises in order to further optimize production processes. The subject of research is the design and optimization of structured cabling systems to automate the production processes of the enterprise. The aim of the work is to improve the method of designing and optimizing structured cabling systems for the transport needs of the cellular network for the needs of the enterprise. The study determined that the process of planning a radio network to automate production processes of the enterprise must be performed in accordance with the following sequence of steps: design of radio network coverage to determine the location of each base station (5G - gNb) and construction of communication transport segment with location cross-country premises. In this paper, research is conducted on the construction of a communication transport segment with the location of cross-country premises. For this purpose, the scope of the statistical method is determined and the choice of the location of the technical premises of the cross floor is substantiated. Conclusions. The research conducted in this paper allowed to develop methods for improving the network architecture of enterprises in order to further optimize production processes. As part of this, a 5G network planning method was developed to automate the production processes of the enterprise, which is to consistently ensure the design of the radio network coverage to determine the location of each base station using an optimized model to estimate the loss of radio signal power. bandwidth, number of connections and reliability and construction of the communication transport segment with the determination of the optimal location of cross-country premises. The developed method makes it possible to plan the optimal structure of the 5G cellular network to optimize production processes, evaluate and reduce the total cost of building the network, while providing the necessary indicators of service quality of network nodes and its reliability.