The article discusses the developed model for recognizing a clothing brand by image. The model not only predicts the type and brand of clothing, but can also determine their similarity. At the initial stage, a dataset was collected containing images of clothing from various brands with a total volume of 9,000 images. In this work, the ViT (Vision Transformer) neural network architecture was used, a model for working with images, which was presented by experts from Google Brain. The vit-base-patch16-224 model acted as a representative of the transformer architecture. Before training, all images were converted to black and white, and data augmentation was also used: image rotation by a random angle, mirror transformation. All photos have been normalized – pixel coordinates have been adjusted to the interval [0,1].

Keywords: neural network, model, machine learning, Vision Transformer, fashion industry, clothing brand prediction, clothing type prediction, brand similarity determination

The text describes software tools for analyzing the structure of metallic materials, including ferrous and non-ferrous metals. It presents image processing methods for edge detection and segmentation of structural elements on the metal surface. A Python program is described, which applies watershed algorithms and searches for white and black grains to segment metal images. The program performs analysis of grain sizes and shapes, and the results are presented visually and for further use. This tool is crucial for quality control and optimization of the properties of metallic materials.

Keywords: software tool, metal, quantitative analysis of microstructure, computer program, Python programming language

Large industrial enterprises can be compared to a complex dynamic system in which management decisions are constantly required. One of the main management decisions, on which the main performance indicators of the enterprise depend, is the process of managing the planning of production. In the process of organizing decision-making, information systems can be used, which are based on mathematical and heuristic calculation methods.organizations of the new enterprises, redistribution of investments in interests of the organization and development of new production on available floor spaces. The most important organizational economic targets of a diversification of management are presented by innovative activity of the industrial enterprise.

Keywords: production planning, resource allocation, Kanban

The article analyzes the testing time of the control system on the test bench, identifies the components of the testing time, and provides the calculation procedure for a typical test bench. The results obtained can be used to estimate the time of testing control systems at the design stage of test benches.

Keywords: automatic control system, rocket and space technology, test bench, analysis of the testing process, experimental testing, testing time

In this paper, the importance of transport networks for the development of the country's economy is considered. It is determined that at the same time, a huge number of heterogeneous factors affect the functioning of transport networks. In this case, in order to optimize transportation processes, it is necessary to have information about the state of the road network in advance. Geoinformation systems are an effective tool for presenting data of this kind, which allow storing and processing spatial and related attribute data. The author proposes a technique for geoinformation mapping of road conditions using the open OSM project. Openstreetmap provides access to up-to-date geographical information, as well as software tools for data processing. The paper defines the type and levels of damage to the roadway, a geoinformation project has been developed, which reflects the main types of damage to the roadway of a section of the road network.

Keywords: geoinformation system, geoinformation mapping, road network, OpenStreetMap

Maintaining the optimal age structure of the forestry fund is an important factor in the use of forest resources. The purpose of this study was to analyze the age structure of the forestry fund of exploitation forests of the Republic of Karelia. For this purpose, data on the age structure of the forest fund by species groups was collected for 17 central forest districts of the study region. Data sources were forest planning documents. The results of the study showed that coniferous forests predominate in the Republic of Karelia. Deciduous tree species are more widely represented in the southern part of the study region. Deciduous and coniferous forests have different age structures. Young stock, mature timber and overmature forest predominate. At the same time, Young stocks are predominantly represented by coniferous forests. A small proportion of forest approaching maturity is one of the fundamental problems of the region under study, as it helps to curb the increase in logging volumes.

Keywords: forest resources, logging, age structure, coniferous species, deciduous species, ripening forests

Overview of existing methods for diagnosing faults in synchronous electric motors and methods for their detection. Classification and analysis of existing methods, their applicability in detecting faults, advantages and disadvantages. Three classes of possible faults in synchronous permanent magnet motors are considered and described: electrical faults, mechanical faults, and demagnetization. The article discusses three classes of diagnostic methods: based on the construction of a mathematical model of a real electric motor and modeling its errors, based on processing signals from sensors, and intelligent methods based on processing collected data using artificial intelligence. The following error detection methods based on modeling are considered: detection based on the model of the electrical schematic, based on the analytical model, and based on the digital simulation model. The following frequency-time analysis methods of the obtained signals from the sensors are considered: analysis using fast Fourier transform, short-time Fourier transform, wavelet transform, Hilbert-Huang transform, and Wigner-Ville distribution. The following intelligent diagnostic methods are considered: diagnosis using convolutional neural networks, recurrent neural networks, support vector machines, fuzzy logic, and sparse representation.

Keywords: Synchronous motor with permanent magnets, faults of electric motor, modeling, fast Fourier transform, wavelet transform, Hilbert-Huang transform, Wigner-Ville distribution, neural networks, fuzzy logic, support vector machine, sparse representation.

The article discusses the possibilities of implementing the model of subject-language integrated learning in higher education institutions, and the authors focus on their professional experience in implementing this model in Southern Federal University for undergraduates. The authors managed to highlight in detail the working options of CLIL, taking into account the specifics of the modern labor market. The authors conclude that the presented technology can and should be applied in practice while observing the principle of feasibility in the perception of professional content in a foreign language by the target audience. This approach requires the teacher to be flexible in the development of such disciplines, since their subject content is adapted in accordance with the language level of the students.

Keywords: content and language integrated learning, social order, bilingual education, communicative approach, professionally-oriented content, didactic principles, professional

The paper considers a model of a multi-aperture wave-front sensor for an active laser beam control system based on the iterative image reconstruction algorithms with limitations, particularly, on the Gerchberg-Saxton algorithm. The specifics of these algorithms is the presence of the so-called divergence factor which is characterized by obtaining “successful” and “unsuccessful” solutions, and may be clarified by stagnation conditions available (or by local extrema). The use of global optimization methods allows to avoid this constraint and to build quite an effective strategy for retrieving phase information. An experimental research was conducted to restore phase information using this method. For this purpose, a model of a seven-channel laser system with a different phase shift was developed.

Keywords: multichannel laser systems, wavefront sensor, Gerchberg-Saxton algorithm, physical modeling, image reconstruction, phase retrieval

The paper considers the problem of managing the reliability of the departmental segment of the feedback platform, which ensures the reception and consideration of citizens' electronic appeals by the Federal Penitentiary Service of Russia. In this aspect, the reliability optimization problem is an optimal control problem with phase constraints. Using the method of penalty functions, the problem of minimal control is formulated, in which phase constraints are taken into account using an external quadratic penalty function. To solve the resulting problem, a wide range of analytical and numerical methods for finding the optimal solution is available.

Keywords: public administration, feedback platform, departmental segment, queuing system, resource management, optimal management task, optimization task, Federal Penitentiary Service of Russia

This paper discusses the Vue, Svelte, Solid and Lit frameworks used to create the View part of client web applications. Their design, the proposed approach to developing client applications, the problems they solve and create, their strengths and weaknesses, as well as their features and limitations in application are studied.

Keywords: Vue, Svelte, Solid, Lit, Web Components, view frameworks, client web applications, front end, rendering

A method for recording holograms using digital cameras with high spatial resolution is considered. To register holograms obtained in optical setups with an inclined reference beam, a high resolution of registration systems is required. To do this, it is necessary to use media with a resolution of 2000-4000 lines per mm. The use of photographic plates requires a fairly long exposure and development time, which is usually done separately from the optical setup. In the case of holographic interferometry systems, it is necessary to provide for mounting the hologram back into the optical setup with sufficiently high accuracy. Therefore, digital holography methods have been developed to record holograms on photomatrices with limited resolution. These methods are based on the use of optical schemes at small angles (less than 5 degrees) between interfering beams. Recently, sensors with a single element size of 1.33 µm and 0.56 µm have appeared. This resolution makes it possible to return to registration schemes with angles between interfering beams of 30-60 degrees. This allows us to hope for the revival of holographic methods and methods of holographic interferometry at the modern level without the use of intermediate recording media.

Keywords: holography, holographic interferometry, photomatrices with high spatial resolution, holography with an inclined reference beam, digital holography, Fourier transform

The article deals with the problem of inaccurate allocation of the boundaries of ore pieces after an explosion in a quarry in the photo. In this article, the possibility of using neural networks for segmentation of photographs was investigated, and training, testing and comparison of the pre-trained bases of neural networks were carried out. The family of pre-irradiated bases EfficientNet and SEResNet was tested on the FPN neural network. Neural networks were tested on the same number of learning epochs, and competitively on three, five, seven and ten learning epochs. Training for more than ten epochs was impractical, since almost all networks were undergoing retraining. According to the results of the test and comparison, the result was obtained that the FPN neural network on the pre-trained EfficientNetB2 bases after 7 epochs of training has a segmentation quality of 98.93% in three segmentation classes and 55.1% in the "ore pieces" class.

Keywords: segmentation, neural network, pre-trained foundation, EfficientNet, SEResNet

The purpose of this study was to assess the forest exploitation conditions of the central forestries of the Republic of Karelia and the Murmansk region. To do this, data were collected for 27 central forestries of the study region, representing 20 variables characterizing indicators of wood resources, natural production conditions and road infrastructure. The assessment of forest exploitation conditions was carried out on the basis of the developed system of indicators, which includes 5 indicators: the value of the exploitation fund; the representativeness of the territory to the natural conditions of growth of deciduous species and the level of allowable cutting area for softwood farming; representativeness of the territory with favorable natural and production conditions in the study region; size of allowable cuts for coniferous economy; and the quality of the road network. The indicators were factors extracted by the method of principal axes in the process of factor analysis. Factors explained 81.4% of the total variance of the 20 variables initially selected. The results of the study showed that two groups of forest areas are distinguished in terms of the average forest reserve: the forest areas of the Murmansk region, which are characterized by a low average reserve, and the forest areas of the Republic of Karelia. Four central forestries located in the southern part of the Republic of Karelia are distinguished by the level of reserves in the forests of deciduous tree species and the level of allowable cut for deciduous farming. The most favorable natural and production conditions are forest areas located in the southern part of the Republic of Karelia.

Keywords: logging, natural and production conditions, factor analysis, Karelia, Murmansk region, timber resources

The problem of computing symmetric Hadamard matrices of the Balonin-Siberry construction is considered. To obtain such matrices, a large number of random binary sequences are required to select three of them, which are bound by the requirements of the matrix design. Such sequences are the first rows of three cyclic blocks of Hadamard matrices. The background of the emergence of quantum computing and the advantage of quantum generation of binary sequences for subsequent selection are considered. The calculation of Hadamard matrices is proposed as a test problem for quantum computers, which allows to show quantum superiority.

Keywords: quantum computers, qubits, random number generators, orthogonal matrices, Mersenne matrices, Kronecker product

The article is devoted to the analysis of modern data encryption algorithms. The introduction gives an overview of the most common encryption algorithms, such as AES, RSA and SHA. The main part of the article includes an analysis of vulnerabilities of modern encryption algorithms and considers various attack methods. It concludes by drawing conclusions that it is necessary to use comprehensive data protection methods and periodically update the encryption algorithms used to prevent possible attacks.

Keywords: Encryption algorithm, data security, vulnerability, attack method, complex method of data protection

Тhe article discusses the mechanism for the rapid creation and maintenance of Kubernetes clusters without low-level operations with a significant reduction in time and labor costs using automation based on Rancher, VMware vSphere and Ubuntu products.

Keywords: Kubernetes, Ubuntu, Rancher, Docker, cluster, containerization, automation

Forecasting the consumption of electrical energy and power is an urgent and significant problem. This paper discusses current methods for predicting the consumption of electrical energy, reflected in various scientific papers, their analysis is carried out with the identification of more promising forecasting methods.

Keywords: energy consumption forecasting, statistical forecasting methods, neural network forecasting methods, hybrid forecasting methods

The paper considers a data visualization model of numerical simulation of a convective cloud. Data visualization is used for visual representation and analysis of processes in a convective cloud. New methods and approaches to visualization of results based on modern technologies and algorithms, such as real-time visualization and the use of calculations on GPUs. An approach to automating the process of rapid qualitative analysis are presentedThe paper aims to create a specialized software for three-dimensional visualization. This software is used for tasks related to the study of complex processes of interaction of processes in a convective cloud based on a numerical model with detailed microphysics.The article considers a data visualization model for numerical simulation of a convective cloud. Data visualization is used for visual representation and analysis of processes in a convective cloud. New methods and approaches to visualization of results based on modern technologies and algorithms, such as real-time visualization and the use of calculations on graphics processors, are presented. The approach to automating the process of rapid qualitative analysis of numerical simulation data is to create specialized software for three-dimensional visualization. This software is used for tasks related to the study of complex processes of interaction of processes in a convective cloud based on a numerical model with detailed microphysics.

Keywords: application program, three-dimensional visualization, numerical simulation data, machine graphics, visualization model, cloud parameters, modeling

The article is devoted to the creation of a specialized automated information system for accounting project decisions in the structural unit responsible for this business process in companies engaged in project activities. Targeted development of such software products will cover and take into account all features of engineering organizations. In turn, electronic accounting of such documentation will not only simplify and accelerate the work of employees, allowing them to pay attention to their other responsibilities, but also reduce the human factor in such a responsible process, when the documents of project decisions are the core of activity.

Keywords: automated information system, design decision accounting system, sequence diagram, precedent, database, report, Windows Forms

The operation of unmanned aerial vehicles (UAVs), as well as their compactness and ease of use, allow them to be actively used in various conditions and situations. UAVs have a number of parameters, from remotely piloted to fully automated, including the degree of their autonomy, as well as design and purpose. AIS transponders receive and process AIS signals, receiving at the output messages about the name of the vessel, data on the course, speed and current navigation status with the display of its location on an electronic map. The use of an open data transfer protocol in AIS makes it vulnerable when false signals (LS) appear in traffic containing distorted information with the display of false targets on the screens of AIS monitors, which can lead to navigation accidents. The purpose of the article is to develop proposals for the detection and identification of AIS UAVs. The paper analyzes the possibility of the formation of an AIS UAV, considers approaches to solving the problems of detecting and identifying drugs and suggests methods for their implementation of technical analysis of radio signals.

Keywords: automatic identification system, false signals, false signal detection, bearing

The article considers a probabilistic-temporal model for assessing the intelligence protection of a group of unmanned aerial vehicles for military purposes from enemy radio and radio intelligence complexes in service with the US Army, provides analytical expressions and a calculation sequence with examples.

Keywords: intelligence security, a group of unmanned aerial vehicles, the probability of temporary contact, the probability of energy detection, stealth, signal-to-noise ratio, surveillance reconnaissance zone, detailed reconnaissance zone, radio and radio intellig

Nowadays, the organization security against cyber-attacks is a matter of great importance and a challenging area, as it affects them financially and functionally. Novel attacks are emerging daily, threatening a large number of businesses around the world. For this reason, the implementation and optimization of the performance of Intrusion Detection Systems is an urgent task. To solve this problem, the scientific community uses deep learning methods. In this paper, we pay special attention to attack detection methods built on different kinds of architectures, such as multilayer perceptron, gated recurrent unit, long short-term memory network, recurrent neural network, and convolutional neural network. To train and test their models, we used dataset UNSW-NB 15. The Australian Centre created this dataset for Cyber Security. It created to generate traffic, which is a hybrid of normal and attack activities. In finally we summarize this paper and discuss some ways to improve the performance of attack detection under thoughts of utilizing deep learning structures.Nowadays, the organization security against cyber-attacks is a matter of great importance and a challenging area, as it affects them financially and functionally. Novel attacks are emerging daily, threatening a large number of businesses around the world. For this reason, the implementation and optimization of the performance of Intrusion Detection Systems is an urgent task. To solve this problem, the scientific community uses deep learning methods. In this paper, we pay special attention to attack detection methods built on different kinds of architectures, such as multilayer perceptron, gated recurrent unit, long short-term memory network, recurrent neural network, and convolutional neural network. To train and test their models, we used dataset UNSW-NB 15. The Australian Centre created this dataset for Cyber Security. It created to generate traffic, which is a hybrid of normal and attack activities. In finally we summarize this paper and discuss some ways to improve the performance of attack detection under thoughts of utilizing deep learning structures.

Keywords: network traffic, computer attack, artificial neural network, traffic analysis, neural network configuration

The article discusses commodity futures price forecasting using a temporal convolutional network. Commodity futures forecasting is an important task for investors and traders because it allows you to predict future prices and the direction of the market. Commodity futures forecasting can be done using a variety of methods and approaches. One such approach is the use of deep learning models, which consists in predicting futures quotes using artificial neural networks. There are many types of neural networks, among them the most popular for the task of processing time series are recurrent neural networks. However, recurrent neural networks have certain disadvantages that a temporal convolutional network does not have. The temporal convolutional network architecture has unique features such as parallel processing of data, extraction of short- and long-term dependencies, and extraction of important features on different time scales. An experiment was conducted to assess the accuracy of predicting the closing price of seven commodity futures using a temporary convolutional network and an ARIMA statistical model with automatic selection of parameters. As a result of the experiment, it was revealed that the temporary convolutional network is superior to the statistical ARIMA model and is a very effective model for forecasting commodity futures. However, despite the high potential of the proposed forecasting model, it is also important to take into account various other analytical methods, such as fundamental analysis and expert opinion.

Keywords: machine learning, temporal convolutional neural network, commodity futures forecasting, commodities, financial time series

Prompt adjustment of the composition of the asphalt concrete mixture is key to achieving high quality asphalt concrete. To enable easy and rapid adjustment of the asphalt concrete mixture formulation, predicting the properties of asphalt concrete (Marshall stability) is critically important. There are many methods for predicting the properties of asphalt concrete, but the choice of one method or another is a very pressing problem. This article proposes two methods for forecasting Marshall stability: forecasting using a multiple linear regression model and forecasting using an autoregressive model. To evaluate the forecasting accuracy of models, we use two metrics: average absolute error (MAE) and average absolute percentage error (MAPE). The results of the study show that the autoregressive model exhibits better forecasting results, especially the second-order autoregressive model.

Keywords: asphalt concrete, control, composition adjustment, forecasting, multiple linear regression model, autoregression model, Marshall stability, forecast accuracy, mean absolute error, mean absolute percentage error