Convenient Graphical Interface Research Articles

Гибридная интеллектуальная система машинного обучения для моделирования процессов обработки фосфатного рудного сырья

The results of a study are presented, the purpose of which was to create an intelligent machine learning system for modeling the processes of charge agglomeration during processing of phosphate ore raw materials. The relevance of the study is justified by the need to improve the information support of technological systems management processes in the context of the digital transformation of the production environment, carried out within the framework of the Fourth Industrial Revolution and characterized by the massive introduction of the industrial Internet of things, which leads to an avalanche-like increase in the volume of technological data. Their processing using modern analysis methods, including artificial intelligence methods, can improve the quality of decisions made and provide competitive advantages. The scientific novelty of the research results is the structure of the proposed hybrid intelligent machine learning system for modeling phosphate ore processing processes, which is based on the joint use of a dynamic model of the sintering process in the Simulink environment and a deep neural network. The architecture of the neural network was developed taking into account the specifics of the mathematical description of the agglomeration process and includes input fully connected layers that receive measurement results of process variables, as well as a recurrent layer that processes the combined sequence from the outputs of fully connected layers. The integration of a Simulink model and a deep neural network makes it possible to quickly adapt an intelligent system to a specific sintering machine through the use of a two-stage machine learning procedure – first on a Simulink simulation model, and then on a real object. Taking into account the significant inertia of the processes accompanying agglomeration, this approach ensures prompt changes in the settings of the hybrid intelligent machine learning system for the new composition of raw materials and technological parameters. A program has been developed that provides a convenient graphical interface for preparing and using an intelligent system, and simulation experiments have shown that the process of additional training for new technological parameters is much faster than initial training while maintaining high accuracy of the obtained modeling results.

WEEGIT – software for visualization and annotation of the electrophysiological activity registration data

Wizard for EEG Information Tabling – “WEEGIT” is a software designed for visualization and annotation of the long-lasting records of activity. Long-lasting records of raw data are used in various areas of life sciences and well represented in neuroscience. In spite of development of other techniques and approaches the electrophysiological activity registration remains the “golden standard” in neurobiological research. Therefore, we developed and verified “WEEGIT” as a powerful and convenient tool for electrophysiological recordings’ description. We combined the most important features presented in other commercial and noncommercial software in the “WEEGIT”. Presented software can operate with widely used formats of records. ”WEEGIT” allows adaptive visualizing of up to 512 channels of record in different timescale without loss of efficiency and consumption of large machine resources. Visualizing also includes optional built-in temporal-spatial analysis (density of current sources or the density of action potentials) displayed as a background image. Integration of a set of optional filtering and signal transforming procedures allows improving record visualization. “WEEGIT” has a convenient graphical user interface with opportunity of simultaneous time browsing and annotating of the records in one workspace. Annotation can be done by simple text information typing as well as an interactive placing of specialized labels and objects. Software also allows saving data in the resampled format, both for the whole record and for user-defined events. In conclusion, “WEEGIT” provides a great set of benefits for the convenient workflow for beginners and specialists in the electrophysiological area of research, including preparation of the data for further specialized analysis.

Development of compressive strength prediction platform for concrete materials based on machine learning techniques

With the continuous development of artificial intelligence, machine learning (ML), as an important branch, is used to promote the digitalization of concrete. Considering that the verification of the strength corresponding to the concrete mixture requires strict curing system and cycle, high cost and low efficiency, ML technology is employed to predict the 28-day compressive strength of ordinary concrete for the support of engineering practices. The aim is to develop an efficient platform for predicting the compressive strength of concrete materials. Ridge Regression (RR), Least Absolute Shrinkage and Selection Operator (Lasso), Elastic Net Regression (EnR), Support Vector Regression (SVR) and K-Nearest Neighbor (KNN) are established, of which efficiency and accuracy are analyzed and calculated in detail. Then, integrated algorithms, namely Gradient Boosting Decision Tree (GBDT), eXtreme Gradient Boosting (XGBoost) and Random Forest (RF), are used to improve the prediction accuracy of a single machine learning algorithm. Finally, Grid Search (GS) optimization algorithm is used to further improve the generalization performance of the model. The results show that the R2 in training set of the single models is about 0.8, which is obviously lower than that of ensemble models (>0.95). It is encouraging that after conducting grid search optimization, the prediction accuracy of ML models can be improved obviously, of which R2 increasing by around 15%. In addition, the GS-XGBoost model achieved the most excellent prediction accuracy and generalization performance, with R2 values of over 99% in both the training and test sets. Finally, based on the results obtained, a software platform with a convenient graphical user interface (GUI) has been developed, which can greatly improve the efficiency of material design and testing required by researchers and engineers.

Application of machine learning to guide efficient metal leaching from spent lithium-ion batteries and comprehensively reveal the process parameter influences

Recycling spent lithium-ion batteries (LIBs) has been a global hot spot due to the increasing scrap amount and enriching valuable resources. Acid leaching is the crucial step of hydrometallurgy recycling metals from spent LIB. However, the uncontrollable components cause many trials for obtaining the optimal leaching parameters, which increases the recycling cost and environmental risk. Herein, this study applied machine learning (ML) to guide metal leaching from spent LIBs and could quickly obtain the leaching results without the vast leaching experiments. After referring to the relative studies between 2005 and 2022, a total of 17,588 data points were collected as the dataset for ML. The process parameter influences of metal leaching were comprehensively revealed through the ML analysis of 20 input features and 4 output metal leaching efficiency. Then, a convenient graphical user interface (GUI) was developed to guide efficient metal leaching from spent LIBs, only measuring the particle size and composition of waste and avoiding extensive leaching experiments. Finally, some experiments verified the GUI reliability. This study can significantly reduce the experiment cost and environmental risk, which trailblazes the smart and efficient recycling of spent LIBs.

Intelligent Workflow and Software for Non-Target Analysis of Complex Samples Using a Mixture of Toxic Transformation Products of Unsymmetrical Dimethylhydrazine as an Example.

Unsymmetrical dimethylhydrazine (UDMH) is a widely used rocket propellant. Entering the environment or being stored in uncontrolled conditions, UDMH easily forms an enormous variety (at least many dozens) of transformation products. Environmental pollution by UDMH and its transformation products is a major problem in many countries and across the Arctic region. Unfortunately, previous works often use only electron ionization mass spectrometry with a library search, or they consider only the molecular formula to propose the structures of new products. This is quite an unreliable approach. It was demonstrated that a newly proposed artificial intelligence-based workflow allows for the proposal of structures of UDMH transformation products with a greater degree of certainty. The presented free and open-source software with a convenient graphical user interface facilitates the non-target analysis of industrial samples. It has bundled machine learning models for the prediction of retention indices and mass spectra. A critical analysis of whether a combination of several methods of chromatography and mass spectrometry allows us to elucidate the structure of an unknown UDMH transformation product was provided. It was demonstrated that the use of gas chromatographic retention indices for two stationary phases (polar and non-polar) allows for the rejection of false candidates in many cases when only one retention index is not enough. The structures of five previously unknown UDMH transformation products were proposed, and four previously proposed structures were refined.

Collision avoidance by constructing and using a passing area in on-board controller

The object of research is the processes of automatic optimal passing of one's own ship with many dangerous targets, including maneuvering ones, by the method of constructing the area of permissible passing parameters in the on-board computer. According to the European Maritime Safety Agency (EMSA), the largest number of ship accidents in 2014–2019 occurred due to collision (32 %). On modern ships, for observation and passing with targets, ARPA (automatic radar plotting aid) is used, which allows to automate manual operations, and the built-in function «Playing the maneuver» provides the navigator with a convenient graphic interface for solving passing problems. At the same time, ARPA is an automated system that assumes the presence of an operator in the control circuit. The presence of a person in the control circuit is related to the «human factor», which is a prerequisite for the occurrence of various types of accidents, including ship collisions. The most effective means of reducing the influence of the «human factor» on control processes is the introduction of automatic control modules in automated systems. The paper develops a method for the passing module, which allows automatic and optimal passing with many targets, including maneuvering ones. The number of targets for passing is not limited by the method, but is limited only by the capabilities of the ARPA to track the targets. The obtained results are explained by the fact that at each step of the on-board computer, a region of permissible passing parameters is constructed for all purposes, passing parameters that optimize a given optimality criterion are selected from the constructed region, the selected parameters are used as software in the control law. The developed method can be used on ships, subject to integration into the existing automated system of an on-board computer with an open architecture, to increase the capabilities of automatic traffic control, in this case, the possibility of automatic optimal passing with many objectives, including maneuvering.

Encryption Methods and Algorithms Based on Domestic Standards in Open-Source Operating Systems

The paper describes the principles and methods underlying the creation of an application in secure operating systems, which provides reliable data encryption. The research aims to analyze and indicate the specifics of encryption methods and algorithms based on domestic standards in open-source operating systems. Cryptanalysis was used in the article, as this avoids vulnerabilities identified in previously created implementations. In the article, the authors draw attention to the fact that 7-Zip uses CBC encryption (concatenation of encrypted text blocks), but the Counter Mode is supported. The same support was provided in the encrypt implementation. Since the key expansion function initially fills the special array created by p7zip with round keys using a unique property of the domestic standard, only one round encryption function was created (performed both during encryption and decryption). This method is also used in various modes. In many cases, initialization time deviations depending on the selected mode are insignificant. The created cryptographic module was tested to meet the domestic standard, which contains several test cases. It was confirmed during the tests that the created module implements the algorithm of the domestic standard. The article shows a way to implement a fairly convenient graphical interface for accessing the cryptographic module, which enables the user not to call the command line and remember the sequence and types of parameters passed to p7zip. This implementation also takes into account the verification of the correctness of decryption and the reading of other error codes.

OOPS: A GUI-based software package for object-oriented analysis of fluorescence polarization microscopy data.

Excitation-resolved fluorescence polarization microscopy (FPM) is a powerful technique that enables the study of orientational order and dynamics in biological systems. Compared to methods like X-ray crystallography, nuclear magnetic resonance spectroscopy, and cryo-electron microscopy, FPM offers key advantages, particularly when applied to lipid membranes, membrane proteins, or membrane-associated protein assemblies. Unlike traditional structural methods, FPM can be configured on standard optical microscopes by simply adding a polarization modulator to the excitation path. However, these experimental advantages are countered by the extensive image processing and analysis required to interpret the polarization response data. Consequently, the use of FPM remains limited to a small subset of laboratories, typically relying on custom-built software designed for a particular biological system of interest. Here, we describe OOPS (Object-Oriented Polarization Software), a software package written in MATLAB that offers a convenient graphical user interface, designed to make FPM analysis accessible to researchers at all skill levels. We demonstrate the power and flexibility of our approach by analyzing FPM data acquired of both punctate and filamentous structures. First, we apply OOPS to desmosomes, cell adhesion complexes that are incredibly difficult to study using traditional methods. Our analyses reveal novel, domain-specific differences in orientational order in the transmembrane components of desmosomes, the desmosomal cadherins. We further show how FPM data can be used to reveal the underlying geometry of large, membrane-associated protein assemblies. Finally, to illustrate the flexibility of OOPS, we also collected and analyzed FPM data of F-actin. We demonstrate how our software can quickly and reliably retrieve pixel-by-pixel F-actin filament orientations under a variety of experimental conditions. Importantly, OOPS does not rely on a specific labeling method, meaning the approaches detailed here could easily be extended to a wide array of biological systems.

Корпус архивных документов Войска Донского: проблемы морфологического анализа

The article presents the results of the collective project aimed at comprising a special annotated diachronic corpus of documents of the 18 th – 19 th cen. from the "Mikhailovsky Stanitsa Ataman" Archive Fund (State Archive of Volgograd Region, Russia). In the course of the work, linguistic, technical and software tasks related to meta-marking, morphological tagging and representation of marked texts in an electronic search environment were solved. The texts are written in cursive script of the 18 th cen. with the use of the old Cyrillic letters, which have spelling specificity. To work correctly with them, an add-on to the stemming tool MyStem by I. Segalovich was created. This application adds to the MyStem the following capabilities: the option to work with the old Cyrillic symbols, a convenient graphical interface; it provides the opportunity to remove homonymy manually, enables marked text exporting to an external data storage and processing system. Morphological analysis of some texts revealed the presence of nominal case form variants, which were not noted in the "Russian Grammar" by M.V. Lomonosov, in modern studies of literary texts of the 18 th century. These findings point to effectiveness of automatic tagging which allows word form correction. The research results substantiated text tagging software tools adjustment for the extension of homonymous forms grammatical analysis options, aimed at identification and manual removal of homonymy. A quantitative analysis of these variants will allow the authors to evaluate their significance for the regional administrative language. The information obtained confirms the importance of the corpus creation for studying the history of the Russian language.

Application of Group Method of Data Handling on the Ultimate Conditions' Prediction of FRP-Confined Concrete Cylinders.

Fiber-reinforced polymer (FRP) is widely used in the field of structural engineering, for example, as a confining material for concrete. The ultimate conditions (i.e., compressive strength and ultimate axial strain) are key factors that need to be considered in the practical applications of FRP-confined concrete cylinders. However, the prediction accuracy of existing confinement models is low and cannot provide an effective reference for practical applications. In this paper, a database containing experimental data of 221 FRP-confined normal concrete cylinder specimens was collected from the available literature, and eleven parameters such as the confining stress, stiffness ratio and strain ratio were selected as the input parameters. Then, a promising machine learning algorithm, i.e., group method of data handling (GMDH), was applied to establish a confinement model. The GMDH model was compared with nine existing models, and the prediction results of these models were evaluated by five comprehensive indicators. The results indicated that the GMDH model had higher prediction accuracy and better stability than existing confinement models, with determination coefficients of 0.97 (compressive strength) and 0.91 (ultimate axial strain). Finally, a convenient graphical user interface (GUI) was developed, which can provide a quick and efficient reference for engineering design and is freely available.

Decision-Making Support Software Tools Based on Original Authoring Bayesian Probabilistic Models

The article presents programs that implement original approaches to group expert rating estimation and fuzzy inference. They implement probabilistic models based on Bayes’ Formula, previously proposed and published in the works of the authors. In these models, the estimated input data are interpreted as evidence in favor of one or another hypothesis from the set of possible ones, determined by the specifics of the model. All the evidence obtained is, in one way or another, transformed into a set of Bayesian conditional probabilities calculated under the assumption that the corresponding hypothesis is true, and the posterior probability distribution on the set of these hypotheses is used as the output. This distribution is used either directly as a result for decision making, or as a basis for calculating the final result. The features of the software implementation of models on the Java platform are discussed, the advantages of the models, confirmed or identified in the process of software implementation, are noted. The developed programs have a convenient graphical user interface and can be used as decision-making support tools to solve applied problems in the field of expert rating estimation and fuzzy inference.

Membrane Mixer: A Toolkit for Efficient Shuffling of Lipids in Heterogeneous Biological Membranes.

Molecular dynamics (MD) simulations of biological membranes have achieved such levels of sophistication that are commonly used to predict unresolved structures and various properties of lipids and to substantiate experimental data. While achieving sufficient sampling of lipid dynamics remains a major challenge, a commonly used method to improve lipid sampling, e.g., in terms of specific interactions with membrane-associated proteins, is to randomize the initial arrangement of lipid constituents in multiple replicas of simulations, without changing the overall lipid composition of the membrane of interest. Here, we introduce a method that can rapidly generate multiple replicas of lipid bilayers with different spatial and conformational configurations for any given lipid composition. The underlying algorithm, which allows one to shuffle lipids at any desired level, relies on the application of an external potential, here referred to as the "carving potential", that removes clashes/entanglements before lipid positions are exchanged (shuffled), thereby minimizing the energy penalty due to abrupt lipid repositioning. The method is implemented as "Membrane Mixer Plugin (MMP) 1.0" in VMD, with a convenient graphical user interface that guides the user in setting various options and parameters. The plugin is fully automated and generates new membrane replicas more rapidly and conveniently than other analogous tools. The plugin and its capabilities introduced here can be extended to include additional features in future versions.

Automated performance analysis tools framework for HPC programs

In the high performance computing (HPC) field, computer and natural scientists work together to solve complex problems. While computer scientists try their best to meet the needs of domain experts, the latter ones still require deeper know-how in informatics to bring their science onto machines in an optimal way. Applications typically require significant optimisation and tuning efforts to harness the performance capabilities of HPC systems. A wide range of very different analyse tools are available for this purpose, but their use is time-consuming and thus also a financial burden. This work is the first to introduce an extensible framework which does not only provide a convenient graphical interface for HPC performance analysis tools, but also simplifies their usage extremely. On this path the foundation for automated scientific software analysis and optimisation workflows is layed, which is believed to become increasingly necessary.

DISTRIBUTED INFORMATION SYSTEM FOR WATER SUPPLY USING PREDICTION METHODS TO PREVENT ACCIDENTS

The purpose of the study is the development and implementation of an information alarm system for the water supply system. The article describes the developed information system for efficient water supply based on the parameters obtained from devices located on pipes. A conceptual, logical and physical model based on the description of the subject area has been developed. A water supply network database was developed using the MySQL Workbench tool. An interface has been created for monitoring the state of pipelines using the Angular framework and the programming languages TypeScript and JavaScript. The information system allows the user to view information from humidity and flow sensors. The application predicts the humidity around the pipes and the water flow for the subsequent period and receives the appropriate message if an accident is possible. The program provides the user with a convenient graphical interface for tracking possible accidents in the water supply network with reference to the coordinates of the pipes, so that the user can make a timely decision to carry out repairs and prevent such a situation.

Integration of MatLab and R with high-level languages using C# and Microsoft Visual Studio as an example

The paper discusses the technology of integration (providing a single interface of interaction) of different software systems working in a single environment. The levels of integration are indicated, the composition of subtasks (subsystems), which can be solved efficiently using various computing environments, are highlighted. The expediency of joint use of MatLab and R systems with high-level programming languages C ++ and C # is shown due to the limited or lack of tools for creating a convenient graphical user interface for mathematical systems, as well as the weak adaptability of programming languages for mathematical or statistical data processing and solving modeling problems. The aim of the article is to justify the most appropriate technology for integrating MatLab and R with highlevel programming languages to ensure high interaction efficiency and a user-friendly interface for complex mathematical or statistical data processing problems. The analysis of the possibilities and ways of solving the organization of interaction of MatLab and R systems with high-level languages carried out on the example of C # and the Microsoft Visual Studio environment. The possibilities of interaction of the R system and the MatLab system with highlevel programming languages, respectively, are considered. An example of software implementation of the integration of R and MatLab with a C# application is given. The capabilities of the developed program as a whole and the functionality of each of the subsystems used (R, MatLab and applications in C #) are indicated. The most preferable variant of integration of the considered systems - with the use of .NetFramework platform - is singled out. The advantages of using the Common Language Runtime for the implementation of the considered option of integrating R, MatLab and C # applications are noted.

Modelling and simulation of a photocatalytic reactor at high TiO2 concentrations

AbstractThis work presents a model of an annular batch slurry photocatalytic reactor using titanium dioxide at high concentrations and pH close to 7. The model considers the effects of the agglomeration of catalyst particles that take place when the reactor operates close to neutral pH. It also includes the adsorption/desorption equilibrium of the contaminant on the catalyst during 1 h in darkness, followed by the photochemical reaction under UV irradiation. The reactor was modelled by a system of differential equations solved using the Micro‐Cap 12 program. This powerful software package numerically solves the differential equations as an equivalent network, providing a convenient graphical interface for model programming and plotting of results. The model was validated with measurements at several catalyst concentrations from 1–2.5 g/L. Aeroxide P25, supplied by Evonik, was used as catalyst, with UV irradiation at a wavelength of 254 nm. Orange II, a widely used azo dye, was chosen as the model contaminant. The model results were in satisfactory agreement with the experimental data. Simulations were carried out with different initial conditions, to analyze the influence of the system variables on the performance of the reactor. In this way, the optimal design criteria at high catalyst concentrations were found. The mathematical model is simple and robust and can be applied in a straightforward way to the design and scaling of photocatalytic reactors. It is also particularly attractive for educational purposes.

Gravity inversion of 2D fault having variable density contrast using particle swarm optimization

ABSTRACTA Matlab‐based optimization algorithm is introduced for inverting fault structures from observed gravity anomalies. A convenient graphical user interface is also presented for incorporating the input parameters without any technical complexity to any users. The inversion code uses particle swarm optimization, and all control parameters are tuned initially for faster convergence. There is no requirement of prior choice of an initial model, that is the advantage of using global optimization. The optimization technique is versatile enough to handle any depth‐varying density distributions. The maximum number of iterations and stopping criterion is fixed initially for getting the best optimized solution. The inverted model's output in terms of fault structure, observed and inverted gravity anomalies and dip, and vertex location of fault plane can be viewed in the graphical user interface at the end of the optimization process. The optimization algorithm is applied to different synthetic models with fixed and depth‐varying density contrasts. All synthetic models are further contaminated with white Gaussian noise for sensitivity analysis, and detailed uncertainty appraisal was also performed for the reliability estimation. Finally, the optimization is implemented for fault structure inversion of the Aswaraopet boundary fault, India, and found that the optimized solution provides a good agreement with the previously published literature. Optimized results indicate that this novel optimization approach demonstrates a robust implementation of fault inversion for any depth‐varying density distributions.

An intelligent system for transforming natural language queries into SQL and its execution

In today’s world information storing and retrieval plays an important role. Database systems play a key role in the new commercial system for information storage. For accessing the data from database, a person who is not having the knowledge of SQL may find themselves handicapped while dealing with the database. This presents a significant limitation in a developing country such as India, because even today, a very large majority of the population does not have the technical knowledge of how to deal with database systems.In such a case, Natural language processing (NLP) assumes a significant job. NLP is getting perhaps the most dynamic strategies utilized in Human-PC Interaction. It is a part of AI which is utilized for Information Retrieval, Machine interpretation, and Language Analysis. This paper, proposes a system which allows the user to query the database in a compatible mode language, through a convenient Graphical User Interface which results in the data required by the user.

SEDA: A Desktop Tool Suite for FASTA Files Processing.

SEDA (SEquence DAtaset builder) is a multiplatform desktop application for the manipulation of FASTA files containing DNA or protein sequences. The convenient graphical user interface gives access to a collection of simple (filtering, sorting, or file reformatting, among others) and advanced (BLAST searching, protein domain annotation, gene annotation, and sequence alignment) utilities not present in similar applications, which eases the work of life science researchers working with DNA and/or protein sequences, especially those who have no programming skills. This paper presents general guidelines on how to build efficient data handling protocols using SEDA, as well as practical examples on how to prepare high-quality datasets for single gene phylogenetic studies, the characterization of protein families, or phylogenomic studies. The user-friendliness of SEDA also relies on two important features: (i) the availability of easy-to-install distributable versions and installers of SEDA, including a Docker image for Linux, and (ii) the facility with which users can manage large datasets. SEDA is open-source, with GNU General Public License v3.0 license, and publicly available at GitHub (https://github.com/sing-group/seda). SEDA installers and documentation are available at https://www.sing-group.org/seda/.

РОЗРОБКА СИСТЕМИ РОЗПІЗНАВАННЯ ОБЛИЧЧЯ ЛЮДИНИ У ВІДЕОПОТОЦІ З ДОПОВНЕНОЮ РЕАЛЬНІСТЮ

In the paper, the problem of face recognition in a video stream with augmented reality is considered. The current state of this problem is investigated. The general process of face recognition and the basic concepts of augmented reality have been studied. The analysis of modern approaches to solving the face recognition problem is carried out, the strengths and weaknesses of the methods used have been found. A search is carried out for a method invariant to scaling, scene changes, head turns, changes in lighting, accessories, and changes in emotions. An algorithm, architecture, and the soft-ware system that solves the problem of face recognition in a video stream with the elements of aug-mented reality have been developed. A histogram of oriented gradients (HOG) is chosen as the method for face detection; face recognition functionality is developed on the basis of the convolutional neural network architecture – ResNet34. Experimental studies are carried out, the system has been tested on both one and several faces simultaneously. Estimate methods of the recognition quality for the developed software system are determined – the plotting of ROC-curves that show the dependence of the number of false positives on the detection accuracy (true positive rate) and measuring AUC. AUC =0.95 has been achieved during recognition of one face, and AUC = 0.83 – during recognition of sev-eral faces (maximum 4). Statement of the problem. Investigation and analysis of existing approaches to building face-to-face recognition technology in augmented reality systems by analyzing models, methods and algorithms for human face recognition, identifying strengths and weaknesses of existing solutions, choosing the best combination of detection and recognition methods. Analysis of recent research and publications. Approaches have been proposed for the formation of biometric face image templates that can be used for biometric verification or face identification. However, all recent facial recognition results have been obtained through the use of deep convolutional neural networks. In the work of Yu. V. Visilter et al., a convolutional neural network with a hash forest (ZNMHL), based on a convolutional network with a hash layer, has been obtained. J. Betty et al. have investigated how different factors influence recognition quality. The purpose of the article is to prove the effectiveness of the proposed approach based on the histogram of directed gradients and convolutional neural network architecture ResNet34 for the problem of face recognition in a video stream in augmented reality systems. Presenting main material. The basic concepts of augmented reality are analyzed. The process of face recognition is described. The description of the software for the solution of the problem is given and mathematical model is developed. The algorithm of work of the program of face recognition program developed by authors is detailed. The architecture of face recognition system in augmented reality video stream is designed. Results. A software system designed to recognize human faces in augmented reality video streams has shown satisfactory results. The application correctly recognizes the face, available in the database, in different conditions of lighting, head rotation, with the presence of accessory ditches, the closure of some parts of the face, changes in emotions, etc., similarly for recognizing multiple faces at the same time. The system has been tested on 520 examples: 4 people separately and together in different combinations under different conditions of lighting, noise, interference, accessories, emotions. Conclusion. Applying a neural network to the ResNet architecture with appropriate settings for detecting and recognizing human faces in augmented reality video streams is a good choice – this method is invariant to scaling, scene changes, head turns, light changes, accessories, and emotion changes. The system is a platform for further development. In particular, it is planned to conduct experimental studies using other methods of face recognition in a video stream and to perform a comparative analysis of the results, as well as to create a more convenient graphical interface of the program and adaptation for the mobile version.