ISSN 0236-235X (P)
ISSN 2311-2735 (E)

Journal influence

Higher Attestation Commission (VAK) - К1 quartile
Russian Science Citation Index (RSCI)

Bookmark

Next issue

3
Publication date:
13 September 2024

Articles of journal № 1 at 2023 year.

Order result by:
Public date | Title | Authors

11. Recognition complexity when developing video monitoring software [№1 за 2023 год]
Author: A.Yu. Kruchinin
Visitors: 1372
The paper considers the problem of choosing an optimal video monitoring mode when using neural network models as a recognizer when different models are more effective on a video stream at different times. Video monitoring tasks are different while the conditions for obtaining data are different, which can be expressed in the recognition complexity concept. Evaluation of the recognition complexity in monitoring allows saving computing resources, thereby reducing the cost of implementation and use. After evaluating the average complexity of recognition, it is possible to choose the optimal recognition mode in terms of speed and relia-bility during post-processing, when time for it is limited. The paper shows the problem solution in the task of two type object detection using YOLOv5 models, when the video stream must be processed in real time with a minimum delay when the result is returned after each frame. The metrics used in the object detection are analyzed in terms of a possibility of assessing the reliability of the results when there is no final information about an object. There is a chosen efficiency crite-rion based on the sum of the F1-score and the cost of computing resources, which makes it possible to eval-uate the model effectiveness for specific objects. The paper shows the dependence of the efficiency criterion on the F1-score for two models. There are the results of testing two models and a dynamic mode based on choosing an appropriate model depending on the input object. The paper describes the limitations of the ap-proach, which can be used only for streaming recognition, when the images received for recognition are only slightly different from the previous ones. in the end, there is a conclusion about the approach applicability for a number of problems in accordance with the restrictions.

12. Vectorization of the three-dimensional immersed boundary method for improving the efficiency of calculations on Intel microprocessors [№1 за 2023 год]
Authors: Rybakov, A.A., Meshcheryakov A.O.
Visitors: 1785
The work is devoted to increasing the efficiency of modern computational applications on high-performance computing systems. The authors consider program code vectorization as a tool for increasing efficiency. Vectorization helps combining scalar operations of the same type into vector analogs, significantly increas-ing performance. Modern Intel microprocessors were chosen as the target platform, for which a unique set of vector instructions AVX-512 is supported. The paper considers an approach to vectorization of a gas dynamic solver using the immersed boundary method and the Steger-Warming upwind scheme in 3D case. This solver has a complex programming con-text that cannot be vectorized automatically. The paper considers the implementation of the solver, as well as approaches to organizing the code and bringing it to a form suitable for automatic vectorization by the icc compiler. To ensure automatic application of vectorization to the solver code, three basic equivalent transfor-mations were applied. First, the calculations, which are the same for all iterations including matrix opera-tions, were localized and brought to the stage of preparing calculations. Second, the main functions of the solver were organized as flat loops, and the data structures were presented as sets of arrays. Third, splitting by condition optimization was applied to loop nests, which can be used to reduce the degree of control branching inside the loop body. These transformations allow the compiler to automatically apply code vec-torization. As a result of the work performed, the solver was accelerated by a factor of 3 due to vectorization when performing calculations on double-precision real numbers.

13. Automated solution of the problem of detecting industrial objects on a orthomosaic using a neural network [№1 за 2023 год]
Authors: Nazarov T.R. , Mamedova N.A.
Visitors: 2991
The task of object detection is common for any image type; only the detection quality indicators and the process technological support of differ. The basis for this study was the materials of aerial photography of industrial facilities. Office processing of aerial photography images by overlapping the original photographs to obtain orthophotomaps of the captured area is a complex and costly process, its automation is fragmen-tary. This article contains a solution for automating the stage of constructing the industrial object contours as a part of the process of their detection on an orthophotomap. As an approach that provided automation, the authors used modeling of a trained convolutional neural network with a one-stage pass through the SSD al-gorithm and based on the backpropagation method. The result was the development of a software package capable of identifying and classifying several objects on an orthomosaic. For optimal work with the software package, there are described types and methods of storing generated data, as well as the implementation of the process of a transition from an image coordinate system to a spa-tial coordinate system using orthomosaic mapping files. The practical significance of the results is in the fact that all steps in the development of the software package are described: there is the given rationale for choosing technologies and algorithms, there are identi-fied and formulated requirements for the software package; there is a described process of training a neural network and the project structure. This allows not only reproducing the proposed solution of the automation problem, but also scaling it considering the input parameters for detecting industrial objects on orthomosaics.

14. Integrated development of software and hardware components of the frequency converter filter for operating in power-limited ship networks [№1 за 2023 год]
Authors: Chernyi, S.G., Sobolev A.S.
Visitors: 2146
Complex energy systems use a large number of different consumers with a non-linear load. Most of them are complex technical and expensive devices. Due to the non-linearity of the load of these consumers, elec-tromagnetic interference occurs in the network, which negatively affects the correct operation of such equipment. Due to the impact of electromagnetic interference on the operation of complex technical equip-ment operating in large power systems, a decrease in the quality of signals and power, as well as a noticeable decrease in the service life of devices affected by interference, can occur. In frequency converters, due to electromagnetic interference in low-current control signals of power elements, there is a possibility of a short circuit, which will lead to a complete failure of the device. To reduce the effect of electromagnetic interfer-ence on devices, passive or active filtering is often used. This article discusses the development and design of an experimental sample of an active filter to use in low-current circuits of frequency converters, as well as the development of an algorithm for the operation of an active filter control system. The research method is to review existing solutions in order to understand how the algorithms are implemented on other devices. The control system, the algorithm was developed for, is implemented on the debugging platform STM32F411E-DISCO. Such solution will make it possible to meet all the emerging needs that arise during the implementation of the active filter device. The creation of the operation algorithm for this filter was carried out in the MATLAB software package. Due to this solution, it becomes possible to create an operation algorithm without additional difficulties asso-ciated with the further translation of the algorithm into C++ code loaded into the microcontroller.

15. The procedure of automatic radar-less speed measurement of mobile objects using a stationary camera [№1 за 2023 год]
Authors: Epifanov V.A., I.O. Temkin, Kalgov I.V.
Visitors: 1743
This paper considers the problem of automatic road scene marking to determine the speed of objects by the radar-less method based on data from only one stationary camera. In addition to the limit in the number of used cameras (no more than one camera per scene), the solution is required to be able to automatically calculate the road markings on a microcomputer. To obtain correct mark- ings, we apply calculating of vanishing points based on the analysis of information about moving road scene participants and calculating of the top view transformation matrix for a scene. Scene marking is defined as a set of virtual lines on a roadway, which allow determining the speed of a vehicle when it is crossing these lines consecutively. The combined use of vanishing points calculation and the construction of the bird's-eye view makes it possible to obtain the required result with minimum computational cost and with sufficient accuracy, despite the problems caused by the image perspective. The paper shows how to apply the approach for automatic marking of road scenes to determine speed on different types of devices, such as x86 platforms and nvidia jetson microcomputers. A distinctive feature of the method is the full automation of the road marking, in which all the stages are implemented automati-cally without operator's help. In addition, the algorithm does not require any pre-calculations related to the characteristics of a camera for taking images. The possibility of deploying the proposed solution on microcomputers allows scaling different systems of monitoring and analyzing road infrastructure. To add a new location to the system it is enough to install a single device, which includes both an input device (camera) and a calculator (microcomputer).

16. A control panel generation module for web projects [№1 за 2023 год]
Authors: Zhukov S.V. , Suslin A.A. , Kovaleva O.A.
Visitors: 1373
The article considers the problems of creating a control panel for web projects. It describes the implementa-tion of a program module written in the php programming language, which allows creating an admin panel for web resources. The module does not require skills in HTML template layout due to the Gentelella Admin template preinstalled in the module (developed and maintained by Colorlib and AigarsSilkalns). The de-scribed module can be used both in projects built on the basis of the popular MVC model and in resources written on the basis of the PHP framework. The control panel includes many standard controls (input fields, switches, tables, lists, etc.), and can also support custom components that add new controls. The purpose of the work is to optimize the process of developing web projects using the developed mod-ule. The subject of the research are ways to implement the administrative panel of a web application. The presented module allows generating html, css and js code for visualizing the control panel. The user interface is set in the form of php code. The objects created in it are interface elements that grouped into forms and panels that a developer needs. Using the module allows reducing the development time spent on the forming the control panel, and also allows creating applications without skills in layout and organization of the client side of the user interface. The theoretical significance of the work is in the analysis of typical elements of the admin panel, the de-scription of the principles of operation of the created module for generating the admin panel. The practical significance is to reduce the development time of web projects using the control panel generation module.

← Preview | 1 | 2