РУС ENG

№ 4(35)'2024

About

News
Aims and Scope
Founder and Publisher
Editorial board
Licensing terms
Privacy Statement
Plagiarism policy
Publication ethics
Archiving Policy
Subscription


For Authors

Instructions for authors
Reviewing proccess
Copyright Notice
License agreement
Article processing charges


Archive

All issues
Search


Contacts

Contacts


Section
ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING


T. V. Yermolenko, R.S. Khakimov
ON THE QUESTION OF APPLICATION OF DEEP LEARNING FOR THE PROBLEM OF CROSS-VIEW GEOLOCALIZATION.

UDC 004.932.2
DOI 10.24412/2413-7383-2024-4-4-15
Language: Russian
Annotation: The article considers the problem of cross-geolocation from a mathematical point of view, highlights the main approaches, problems and key features of the application of deep learning. Problems such as 1) similarity of key objects in detail and uniform appearance are indicated; 2). image styles can vary greatly; 3) the average pool used in convolutional neural networks ignores the interaction between local objects. Among the key features of the application of deep learning are: 1) Transformer-based architectures make it possible to more clearly identify characteristic objects and ignore background information; 2) the results of a Transformer-based model trained twice will have a large spread; 3) the scale of satellite images is fixed, while the altitude of the UAV flight changes dynamically.
Keywords: cross-view geolocation, computer vision, machine learning, neural networks, UAVs.


A. V. Nicenko, V. Ju. Shelepov
THE USE OF SEMANTIC INFORMATION TO DISAMBIGUATE THE NOMINATIVE/ACCUSATIVE HOMONYMS: AN ELEMENT OF CREATING ONTOLOGY.

UDC 004.89:004.93
DOI 10.24412/2413-7383-2024-4-16-24
Language: Russian
Annotation: The article proposes a method for automatic disambiguation of nominative and accusative cases of nouns using information on semantic relationships between words extracted from a large corpus of texts. The data on semantic relationships are presented as an ontology consisting of a set of semantic triplets or triples "subject-predicate-object". The results are implemented in experimental software for disambiguation.
Keywords: natural language processing, disambiguation, ontology, knowledge graph, semantic triple.


Y. S. Pikalyov
KEY OBJECTS DETECTION AND CROSS-VIEW GEOLOCATION: DATASET ANALYSIS AND METHODOLOGICAL ASPECTS.

UDC 004.932.72
DOI 10.24412/2413-7383-2024-4-25-37
Language: English
Annotation: This work focuses on the challenge of developing a system for cross-view geo-localization using a neural network-based approach. This area is highly relevant, as such systems enable UAVs to navigate complex environments by identifying objects, obstacles, and routes, thereby reducing dependence on operators. Additionally, recognition systems allow UAVs to efficiently collect and analyze data. In the course of this study, an analysis of existing datasets suitable for cross-view geo-localization tasks was conducted, and general principles and requirements for such datasets were identified. The following datasets were highlighted: Objects365, LVIS, VisDrone, DOTA, iSAID, and GeoText. Moreover, the key challenges of the task were outlined, including: 1) image preprocessing; 2) feature extraction; 3) the feasibility of pre-training a base model; 4) accounting for the semantic characteristics of objects, among others.
Keywords: neural networks, datasets, cross-view geo-localization, computer vision, object recognition, data augmentation, autonomous systems.

Section
MATHEMATICAL MODELING, NUMERICAL METHODS AND SOFTWARE SYSTEMS



V. I. Zenzerov
MATHEMATICAL MODELING OF HYDROMECHANICAL PROCESSES IN TRANSMISSION NETWORKS OF MINING EQUIPMENT.

UDC 51-74:004.932
DOI 10.24412/2413-7383-2024-3-36-43
Language: Russian
Annotation: The article presents mathematical models of pressure transmission processes through transmissions of mechanized supports of mine workings. Boundary value problems for systems of differential equations and numerical algorithms for their implementation are considered.
Keywords: algorithm, mathematical modeling, process, parameters, hydromechanics.

L. A. lazebnaya
MATHEMATICAL MODELING AS A WAY TO SIMULATE THE REACTION OF A COAL SEAM TO HYDRODYNAMIC EFFECTS.

UDC 537.874
DOI 10.24412/2413-7383-2024-4-55-64
Language: Russian
Annotation: The article presents deterministic mathematical models of the process of hydraulic action on a coal seam, which are based on the equations of nonlinear elastic filtration of liquid in a continuous medium. It is shown by mathematical modeling that the process of humidification of a coal seam under hydraulic action through a single well causes a high level of variation in the increase in humidity in the treated massif.
Keywords: algorithm, mathematical modeling, process, parameters, simulation.


E. V. Perinskaya
MATHEMATICAL MODELS AND COMPUTATIONAL ALGORITHMS FOR THE STUDY OF NONSTATIONARY THERMODYNAMIC PROCESSES IN A CONTINUOUS MEDIUM.

UDC 537.874
DOI 10.24412/2413-7383-2024-4-65-74
Language: Russian
Annotation: The paper considers mathematical models of processes based on boundary value problems for equations of mathematical physics, and ways to solve problems based on finite difference methods. To improve the parameters of technologies, it is necessary to conduct a study of processes, the most effective research method is mathematical modeling.
Keywords: physical process, mathematical model, boundary value problem, algorithm, numerical method.

Section
SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING, STATISTICS




A. I. Borovikov, O. A. Krivodubsky
INTERACTION OF THE MAIN FLOWS OF THE ENTERPRISE.

UDC 519.8:658.512
DOI 10.24412/2413-7383-2024-4-76-87
Language: Russian
Annotation: The article analyzes the interconnection of material, information, and financial flows in supply chains. It emphasizes the importance of their coordination to enhance business efficiency, prevent financial violations, and ensure sustainable development. Special attention is given to the role of automation and information systems in improving logistics.
Keywords: supply chain management, material flows, information flows, financial flows, flow alignment, logistics, automation, fraud prevention, transparency.


E. I. Mashchenko, D. K. Karpov, O. O. Varlamov, L. E. Adamova, E. G. Balenko
CREATION OF A MIVAR EXPERT SYSTEM FOR UNDERSTANDING IMAGES AND MAKING DECISIONS WHEN PEOPLE FALL.

UDC 004.891.2
DOI 10.24412/2413-7383-2024-4-88-100
Language: Russian
Annotation: An image understanding system has been developed as a web application with a comprehensive intelligent video analysis system based on a mivar expert system (MES). The MES is used to determine further actions after detecting a threat of a person falling in a video fragment. The parameters that can be transmitted to the MES from a module based on a neural network and detecting the fact of a fall in a video sequence; contextual parameters that can be obtained directly from the user, as well as parameters that are calculated by the MES itself are determined. For these parameters, the dependencies and rules on the basis of which they are formed were described in detail. Based on the calculated parameters, the MES makes a decision on further actions.
Keywords: mivar expert system, mivar, image understanding, smart video camera, video analytics, web application, neural network, security.


B. V. Pavlenko, Ya. S. Pikalyov
METHODOLOGY FOR CREATING AN AERIAL IMAGE DATASET FOR CROSS-GEOLOCATION TASKS.

UDC 004.8, 004.93
DOI 10.24412/2413-7383-2024-4-101-112
Language: Russian
Annotation: Computer vision systems used in unmanned aerial vehicles (UAVs) play a crucial role in a wide range of tasks, including terrain classification, infrastructure monitoring, emergency detection, and identifying transportation and other objects. However, their application in UAV-based computer vision systems requires comprehensive aerial image datasets. The high costs and restrictions associated with aerial data acquisition under certain conditions pose significant challenges for direct data collection. This study presents a methodology for creating a comprehensive and accessible aerial image dataset using third-party cartographic services such as Google Maps, Google Earth, and Google Earth Studio. An algorithm for automated data collection is proposed, which involves dividing the target map area into fixed-size cells and obtaining two types of images: satellite-view and simulated drone-view. The advantages of the proposed automation approach include time and resource efficiency, though it also highlights challenges related to the dynamic nature of web pages and dependence on the functionality of the selected platform.
Keywords: aerial images, computer vision, image recognition, satellite images, 3D visualization, UAVs, cross-view geo-localization.


V. N. Pavlysh, A. S. Chupakhin
COMPUTATIONAL ALGORITHM FOR ASSESSING THE EFFICIENCY OF PROTECTION AGAINST ELECTROMAGNETIC FIELDS DEPENDING ON THE PARAMETERS OF SHIELDING MATERIALS.

UDC 537.874
DOI 10.24412/2413-7383-2024-4-113-125
Language: Russian
Annotation: The article presents a computational algorithm implemented in the MATLAB environment, which, based on the analysis of electromagnetic and physical parameters of materials, allows determining their efficiency indicators for shielding electromagnetic fields in the frequency range from 50 Hz to 100 kHz. Four variants of domestic materials were studied: two ferromagnetic metal alloys with high magnetic permeability (79NM and 50N), an amorphous alloy (AMAG-170) and one conductor (copper grade M00K). The behavior of materials under various conditions (plane waves, electric and magnetic waves in the near zone) is considered.
Keywords: algorithm, electromagnetic waves, screening, efficiency, material.


N. S. Podoprigorova, S. A. Kozyrev1, S. S. Podoprigorova, A. V. Baldin, A. A. Kotsenko, Gong Shenshuo
DEVELOPMENT OF A MIVAR EXPERT SYSTEM FOR SELECTING A CONSENSUS ALGORITHM FOR DISTRIBUTED LISTS.

UDC 004.89 + 007.52
DOI 10.24412/2413-7383-2024-4-126-138
Language: Russian
Annotation: In this paper, a mivar expert system (MES) was created to determine the optimal consensus algorithm in distributed registries for various user situations and their subject areas. A comparison of 23 algorithms was conducted consensus from 6 categories (graph-based, proof-based, weight-based, Byzantine voting-based, non-Byzantine voting-based, randomized). In the process of analyzing existing consensus algorithms, 15 significant selection criteria were identified, including such as scalability, decentralization, throughput ability, energy efficiency, etc. The Wi!Mi Razumator software package serves as the main tool for creating an MES, which allows choosing appropriate consensus algorithms based on user requests. The MES for choosing a consensus algorithm will be useful for distributed ledger developers when choosing the best algorithm from a variety of available options and taking into account the specifics of each subject area.
Keywords: mivar, mivar networks, mivar expert systems, KESMI, Wi!Mi, artificial intelligence, distributed registries, consensus algorithms.


M. S. Torzhkov, Yu. P. Koroleva, A. V. Baldin, A. A. Kotsenko, Shen Qiujie
CREATION OF A MIVAR EXPERT SYSTEM FOR IMPLEMENTING ETHICAL ASPECTS OF ARTIFICIAL INTELLIGENCE FOR CREDIT SCORING.

UDC 004.89 + 007.52
DOI 10.24412/2413-7383-2024-4-139-150
Language: Russian
Annotation: A mivar expert system (MES) for scoring a credit system with artificial intelligence (AI) has been created. Integrating ethical principles of AI into the process of designing and using scoring models will help make them fairer and more consistent with social values, and will also lead to an expansion of the bank's client base. The created MES assesses the borrower's creditworthiness based on the created rules, which allows increasing the sample of questionnaires for consideration, focusing on parameters that in turn are indicators of ethics and help make the system more transparent, honest and user-oriented. The created ethical mivar expert system increases customer confidence in financial institutions and promotes the development of responsible lending.
Keywords: mivar, mivar networks, mivar expert systems, KESMI, Wi!Mi, artificial intelligence, distributed registries, consensus algorithms.


V.Y. Ustenko, V.I. Bondarenko
SOFTWARE DEVELOPMENT DATA ANNOTATION FOR COMPUTER VISION TASKS: AN OBJECT-ORIENTED APPROACH BASED ON WINFORMS.

UDC 004.8, 004.93
DOI 10.24412/2413-7383-2024-4-151-163
Language: Russian
Annotation: This work focuses on the development of software for data annotation in COCO (Common Objects in Context) and YOLO (You Only Look Once) formats for computer vision tasks. The creation of high-quality annotated datasets directly impacts the performance of machine learning models, making efficient data annotation software a crucial component of research and development in artificial intelligence. The study identifies key criteria for classifying annotation tools, analyzes the capabilities of modern data annotation tools, and develops the architecture and interaction module of the system.
Keywords: annotation systems, software engineering, computer vision, image processing, metadata, machine learning, datasets.


Shen Qiujie, Gong Shenshuo, O. O. Varlamov, L. E. Adamova, E. G. Balenko
DYNAMIC ROBOT PATH PLANNING BASED ON SEMANTIC OBJECT DETECTION USING MIVAR EXPERT SYSTEM.

UDC 004.891
DOI 10.24412/2413-7383-2024-4-164-176
Language: Russian
Annotation: An algorithm for dynamic path planning of a ground robot based on semantic object detection technology is developed to improve human safety when robots operate in large spaces. Our study takes into account the psychological characteristics of human behavior and movement in large spaces. The algorithm combines 2D laser scanning data with information on the categories and positions of objects identified by object detection in images. Depending on the importance of the object category, the algorithm increases the obstacle area on the map, which allows the robot to avoid important objects (e.g., children) when planning a route and improves the safety of the robot in the presence of people. The mivar expert system implements a corresponding prototype of the data fusion algorithm, which can be used in airports, railway stations, shopping malls and other places to improve human safety when robots operate in large spaces.
Keywords: mivar, mivar expert systems, pattern understanding, trajectory planning, object detection, data fusion, obstacle magnification, indoor environment, robot safety.

Section
SYSTEM ANALYSIS, CONTROL AND INFORMATION PROCESSING, STATISTICS



S. I. Ulanov, O. A. Krivodubsky, A. А. Nikitina
ANALYSIS OF MODERN DEVELOPMENTS OF UNMANNED AERIAL VEHICLES.

UDC 528.013
DOI 10.24412/2413-7383-2024-4-178-191
Language: Russian
Annotation: The article examines the main technical and design characteristics of domestic and foreign unmanned aerial vehicles in order to select devices with the most suitable technical and design features for complex control of mobile objects over a long time period.
Keywords: UAV (unmanned aerial vehicle), flight and tactical characteristics (TTX) of an aircraft.

Section
INTERDISCIPLINARY AI RESEARCH. COGNITIVE ASPECT



I. S. Salnikov, R. I. Salnikov
PROBLEMS AND ISSUES OF EMOTIONAL EXPRESSION AND PATTERNING OF EMOTIONS IN THE RESEARCH OF CARROLL IZARD AND HIS PREDECESSORS, CONTEMPORARIES AND FOLLOWERS.

UDC 004.928, 004.93
DOI 10.24412/2413-7383-2024-4-193-209
Language: Russian
Annotation: The review presents the most significant research results of the American psychologist Carroll Izard and in his presentation and interpretation of his predecessors, contemporaries and followers on the problems and issues of emotional expression and patterning of emotions and shows the role of adequate emotion recognition in the formation of emotional intelligence systems.
Keywords: emotional expression, emotion patterning, evolutionary and biological significance of expression, theory of differential emotions, emotional intelligence systems.


I. S. Salnikov, R. I. Salnikov, T. D. Klyushanova
A COMPUTER VERSION OF THE INFORMATION SYSTEM FOR BIBLIOTHERAPEUTIC DRUG-FREE SELF-REGULATION AND VERBAL THERAPY OF PSYCHOEMOTIONAL STATES OF PERSONALITY.

UDC 159.923:159.942.5
DOI 10.24412/2413-7383-2024-4-210-220
Language: Russian
Annotation: The article discusses the possibilities of developing a computer version of an information system for bibliotherapeutic drug-free self-regulation and verbal therapy of psychoemotional states of personality using machine information processing technologies.
Keywords: computer version, bibliotherapeutic drug-free self-regulation, verbal therapy, psychoemotional state of personality, information system.

Section
AUTOMATION AND CONTROL OF TECHNOLOGICAL PROCESSES



S. A. Rylov
SINGLE-PAIR ETHERNET T1S AS THE BASIS FOR INTELLIGENT SENSORS OF THE INDUSTRIAL INTERNET OF THINGS.

UDC 004.324
DOI10.24412/2413-7383-2024-4-222-232
Language: Russian
Annotation: Currently, data digitalization occupies a major niche in the development of industrial automation systems. With the development of the trend of introducing approaches and technologies of the industrial Internet of Things, the transfer of information from the field level in digital form is becoming more and more relevant. This article discusses the possibility of using modern 10Base-T1S data transmission technology as a bus that allows you to implement an intelligent sensor network within local industrial installations and complex units. In addition, the author proposes an architectural concept of an intelligent IIoT temperature sensor with the
Keywords: 10Base-T1S, Industrial Internet of Things, Iot, Smart sensors.