Artificial Intelligence, Intelligence Systems, Neural Networks

PDF

Article # 10_2022

11 p.

PDF

Lyalya U. Bakhtieva, Vladimir M. Bogolyubov, Maxim D. Tumakov
Simulation of a multifunctional micromechanical gyroscope

The possibility of constructing a multifunctional inertial navigation device based on a hybrid-type modulation micromechanical gyroscope is considered. A mathematical model of the device ("heavy" gyroscope) as a high-quality three-dimensional oscillatory system is constructed. It is numerically shown that, under certain conditions, the reaction of the system to the motion of an object has, along with precession, the observed nutation, which carries information about the linear motion of the gyroscope base. It is noted that the possibility of measuring linear accelerations is ensured by the presence of a small symmetrical distance between the axes of the elastic suspension relative to the center of mass of the sensing element. The results obtained make it possible to implement a two-component angular velocity meter and a two-component linear acceleration meter in one device.

Key words: mathematical model, vibration, micromechanical system.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-17-27

Article # 12_2022

15 p.

PDF

Igor V. Vinokurov
Using a Convolutional Neural Network to Recognize Text Elements in Poor Quality Scanned Images

The paper proposes a method for recognizing the content of scanned images of poor quality using convolutional neural networks (CNNs). The method involves the implementation of three main stages.
At the first stage, image preprocessing is implemented, which consists of identifying the contours of its alphabetic and numeric elements and basic punctuation marks.
At the second stage, the content of the image fragments inside the identified contours is sequentially fed to the input of the CNN, which implements a multiclass classification.
At the third and final stage, the post-processing of the set of SNA responses and the formation of a text document with recognition results are implemented.
An experimental study of all stages was carried out in Python using the Keras deep learning libraries and OpenCV computer vision and showed fairly good results for the main types of deterioration in the quality of a scanned image: geometric distortions, blurring of borders, the appearance of extra lines and spots during scanning, etc.

Key words: image processing, convolutional neural network, Python, Keras, OpenCV.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-45-59

Article # 14_2022

18 p.

PDF

Dinara H. Giniyatova, Vasilii A. Lapinskii
Nodules detection on computer tomograms using neural networks

Results of neural networks (NN) application to the problem of detecting neoplasms on computer tomograms of the lungs with limited amount of data are presented. Much attention is paid to the analysis and preprocessing of images as a factor improving the NN quality. The problem of NN overfitting and ways to solve it are considered. Results of the presented experiments allow drawing a conclusion about the efficiency of applying individual NN architectures in combination with data preprocessing methods to detection problems even in cases of a limited training set and a small size of detected objects. Key words: image processing, convolutional neural network, Python, Keras, OpenCV.

Key words: object detection, image processing, neural networks, YOLO.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-81-98

Article # 16_2022

25 p.

PDF

Ilya V. Ketov, Alexander O. Spiridonov, Anna I. Repina, Evgenii M. Karchevskii
Modeling of unidirectional radiation of microdisk resonators with small piercing holes by Galerkin method with accurately computed matrix elements

Using the integral-equation-based in-house guaranteed-convergence numerical code, we study the effect of a circular air hole on the frequency, threshold gain and directionality of emission of the whispering-gallery modes of a two-dimensional model of circular-disk microcavity laser. It is shown that a small hole can enhance the directionality greatly and leave the threshold gain intact, if the disk‘s refractive index is large enough and the hole‘s location is chosen properly. This location should be close to the area, in which the same uniform disk, if illuminated with a plane wave, would display a broadband focusing in the form of a hot spot called a photonic jet. (In Russian).
Key words: software complex, Galerkin method, microdisk laser.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-113-137

Article # 18_2022

13 p.

PDF

Vladimir M. Konyukhov, Ivan V. Konyukhov, Albina R. Ganieva
Calculation of Interrelated Thermal Processes in a Submersible Electric Motor, Rocks and Water-Gas-Oil Flow in a Producing Well

This paper is devoted to the study of interrelated thermal processes in a submersible electric motor of a pumping unit located in an oil-producing well and flowed around by a water-oil-gas reservoir mixture, taking into account the heat exchange of the flow with the rocks surrounding the well. To describe these processes, mathematical and numerical models are developed. The numerical model and algorithms are implemented in a software that allows to study temperature fields and various thermal effects using computational experiments with simultaneous visualization of the results of computations. It is shown, in particular, that the transient thermal processes in the system “motor— three-phase flow – rocks”, when the motor is turned off due to its heating to the maximum permissible temperature depend on the physical and geometrical characteristics of each element of the system and are characterized by a non-trivial temperature profiles in rocks. Calculated estimates of the duration (on the order of tens of minutes) of the cooling stage of the motor after it is turned off and its heating stage when it is turned on again correspond to the real times of these processes in producing oil wells. (In Russian).

Key words: mathematical modeling, finite-difference method, computer simulation, thermal processes, heat exchange, oil-gas-water mixture, oil-producing well, surrounding rocks, submersible pumping unit, electric motor, computational experiments.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-165-177

Article # 20_2022

31 p.

PDF

Vlada V. Kugurakova, Igor O. Antonov, Bogdan V. Goncharenko, Artysh A. Chaibar
Visual digital copy of the crime scene and its presentation in virtual reality as tools in criminal proceedings

The paper presents new possibilities of such technologies as photogrammetry and virtual reality for application in criminal proceedings. The concept of building “on the fly” three-dimensional digital copy of the incident, directly at the scene of the incident is presented. The concept is tested by applying specific technological approaches to create a digital copy of the incident with the server processing the initial information from a set of photos of the scene from several angles, the architecture of the hardware-software system is developed and the functionality of the mobile solution and virtual reality application is described. It is shown how a digital copy of the incident can be used, in particular, in a criminal case by the subjects of proof – both the subjects of the prosecution and the defense. To date, no precedents have yet been found for the use of virtual reality technology in court proceedings in this aspect, and the authors propose a gradual introduction of the proposed approaches into practice. (In Russian).

Key words: photogrammetry, rendering, client-server, virtual reality, VR, criminal proceeding, digital twin, digital copy, crime scene, forensics.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-193-223

Article # 22_2022

14 p.

PDF

Dina S. Latypova, Dmitrii N. Tumakov
Clustering of handwritten digits by Kohonen neural network

Clustering of handwritten digits is carried out for sixty thousand images contained in the training sample of the MNIST database. For clustering, the Kohonen neural network is used. For each handwritten digit, the optimal number of clusters (no more than 50) is determined. When determining the distance between objects (images of handwritten digits), the Euclidean norm is used. Checking the correctness of building clusters is carried out using data from the test sample of the MNIST database. The test sample contains ten thousand images. It is concluded that the images from the test sample belong to the "correct digit" cluster with a probability of more than 90%. For each digit, an F-measure is calculated to evaluate the clusters. The best F-measures are obtained for digits 0 and 1 (F-mean is 0.974). The worst values are obtained for the number 9 (F-mean is 0.903). A cluster analysis is also carried out, which allows drawing conclusions about possible errors in recognition by the Kohonen neural network. Intersections of clusters for images of handwritten digits are constructed. Examples of intersections of clusters are given, as well as examples of images that are incorrectly recognized by the neural network.

Key words: Kohonen Neural Network, Clustering, MNIST.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-241-254

Article # 24_2022

16 p.

PDF

Dmitrii N. Tumakov, Georgy V. Kannunikov, Marat G. Minlebaev
Detecting states of ion channels on the cell membrane using neural networks

The problem of automating the process of analyzing the open states of channels on the membrane of a neuron of a living organism is considered. Taking into account that the registration of the electrical activity of the cell was made by the patch clamp method at various values of the applied potential, a division into intervals with a constant potential is carried out. Further, to eliminate noise, a notch filter, low-frequency and high-frequency Chebyshev filters are applied to the data. A neural network is applied to the normalized data, based on the results of which the data is changed and re-processed by the same neural network. As a result of the algorithm, the dynamics of channel states was obtained, which makes it possible to register up to several open channels simultaneously. (In Russian).

Key words: neural networks, ion channels, detecting states of channels, living organism.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-255-273

Article # 26_2022

18 p.

PDF

Regina A. Sharaeva, Vlada V. Kugurakova, Natalia E. Selezneva
Assessment of time reduction when using a modified task-tracking methodology in IT project management

Task-trackers that allow automating management tasks are traditionally used for IT project development management. Popular tools were analyzed, and new requirements were formulated for task and project management systems in general for any highly specialized areas of IT development. The author’s methodology for task-tracking systems, not found in any of the considered solutions, was developed. Practical implementation of the proposed approach showed that it is possible to solve management problems much more efficiently: optimization reaches more than 50% in some cases. In addition, the developed tool ProjectAR allows leveling several risks.
Comparison with the popular task tracker Asana, which is the closest to ProjectAR by its functionality, was conducted to prove the hypothesis of time reduction for management tasks. In addition to the time metric, the risk of incorrect integration of generated development artifacts was selected as a criterion for tool comparison. The tools were compared based on the number of templates needed to implement IT solutions and the number of typical projects.
At the end, a vision for tool development is given. (In Russian).

Key words: software engineering, IT development, task tracker, automation.

article citation

 DOI

https://doi.org/10.25209/2079-3316-2022-13-3-307-324