Neurovisual recognition of signal radio images

Objectives. The study set out to solve the problem of radiovision classification of objects based on identified features by developing a combined neurovision algorithm for real-time recognition of signal radio images of objects using machine learning (ML) technologies and a fully connected neural network with data augmentation, as well as to improve the probability of correct classification in neurovision signal processing.

Methods. In the study, several methods were used: electrodynamic modeling, machine learning (linear regression, classification, and Random Forest), and deep learning (fully connected neural networks). The bootstrap aggregating (bagging) technique was also employed. An assessment of object classification accuracy metrics and statistical criteria for the reproducibility of radio images was carried out.

Results. A combined neurovision object recognition method was developed that demonstrated a probability of correct classification of at least 0.97 for any of the objects transmitted for training with specified form factors when using augmented data. Data augmentation was shown to increase the neural network’s probability of correct classification by 0.04. The obtained results confirm the adequacy of neural network approaches compared to classical ML methods for neurovision object recognition, particularly when dealing with a limited base dataset of objects for neural network training. The proposed method was tested for basic classification of spherical and cubic object models in the centimeter radio frequency range.

Conclusions. Neural networks with data augmentation demonstrate a probability of correct classification exceeding 0.97 for neurovision recognition of radio images as compared to neural networks without data augmentation (0.04 lower) and traditional ML methods (0.13 lower). Although ML methods are inferior to neural networks in radio image reproducibility, they remain indispensable in cases where computational resources are limited. For real-world applications, database expansion through field experiments and the implementation of hybrid neural network architectures are required.

1. Terletskii A.S., Terletskaya E.S. Neironnye seti i iskusstvennyi intellekt: Osnovy neironnykh setei na yazyke Python (Neural Networks and Artificial Intelligence: Fundamentals of Neural Networks in Python). Lipetsk: Lipetsk State Pedagogical P. Semenov-Tyan-Shansky University; 2023, 76 p. (In Russ.). https://www.elibrary.ru/ugipee

2. Torgaev S.N., Lezhnina I.A., Shul’ga I.D. Prakticheskoe rukovodstvo po tsifrovoi obrabotke signalov: tsifrovye fil’try i obrabotka EHKG signalov (A Practical Guide to Digital Signal Processing: Digital Filters and ECG Signal Processing). Tomsk: STG; 2020, 112 p. (In Russ.). https://www.elibrary.ru/rupfct

3. Tsaregorodtsev M.A. Multithreaded numerical implementation of cryptographic algorithms of parallel action for protecting confidential information in defense-industrial complexes during its processing, storage and transmission to the cloud storage of big data. Al’manakh Permskogo voennogo instituta voisk natsional’noi gvardii = Almanac of the Perm Military Institute of the National Guard Troops. 2023;4(12):104–112 (in Russ.). https://www.elibrary.ru/vvtbii

4. Kurushin A.A. Gibridnoe modelirovanie v HFSS ANSYS (Hybrid Modeling in HFSS ANSYS): A tutorial. Moscow: SOLON-Press; 2023, 292 p. (In Russ.).

5. Bankov S.E., Kurushin A.A. Raschet antenna i SVCh struktur s pomoshch’yu HFSS Ansoft (Calculation of Antennas and Microwave Structures using HFSS Ansoft). Moscow: RODNIK; 2009, 256 p. (In Russ.).

6. Kozhemyako V.A. Getting a signal response from an object in Ansys CAD. In: Actual Problems and Prospects for the Development of Radio Engineering and Infocommunication Systems (“Radioinfocom-2024”): Collection of scientific articles based on the materials of the 8th International Scientific and Practical Conference. Moscow: RTU MIREA; 2024. P. 381–384 (in Russ.). https://www.elibrary.ru/mwzoxe

7. Ivanova V.Yu., Solovyev D.O. Overeview of Big Data Processing Methods Using Apache Spark, Pandas Library, and SQL. Naukosfera = Naukosphere. 2024;5(1):43–47 (in Russ.). https://doi.org/10.5281/zenodo.11241367, https://www.elibrary.ru/uljwcm

8. Habib J.M.T., Poguda A.A. Comparison of Deep Learning Sentiment Analysis Methods, Including LSTM and Machine Learning. Otkrytoe Obrazovanie = Open Education. 2023;27(4):60–71 (in Russ.). https://doi.org/10.21686/1818-4243-2023-4-60-71

9. Kazantsev T. Iskusstvennyi intellekt i mashinnoe obuchenie. Osnovy programmirovaniya na Python (Artificial Intelligence and Machine Learning. Fundamentals of Python Programming). LitRes: Samizdat; 2020, 123 p. (In Russ.).

10. Bolshakov N.I., Sidorova E.V. Comparative Analysis of Machine Learning Methods for Problems of Data Classification. Matematicheskie metody v tekhnologiyakh i tekhnike = Mathematical Methods in Technology and Engineering. 2023;8: 66–71 (in Russ.). https://elibrary.ru/zdgmrk

11. Kouemou G., Opitz F. Impact of Wavelet-Based Signal Processing Methods on Radar Classification Systems Using Hidden Markov Models. In: 2008 International Radar Symposium. Wroclaw, Poland. 2008. https://doi.org/10.1109/IRS.2008.4585763

12. Shevchenko A.S., Samarin V.A. Neironnye seti (Neural Networks): A Tutorial. Moscow: Ai Pi Ar Media; 2025, 181 p. (In Russ.).

13. Smirnov E.E., Kostyleva V.V., Murtazina A.R., Razin I.B. Comparison of Convolutional and Fully Connected Neural Networks in Relation to Image Recognition Tasks. Izvestiya vysshikh uchebnykh zavedenii. Tekhnologiya tekstil’noi promyshlennosti = Textile Industry Technology. Series: Proceedings of Higher Educational Institutions. 2023;5(407): 236–242 (in Russ.). https://elibrary.ru/gvpmpa

14. Paramonov A.A., Nguyen V.M., Nguyen M.T. Multi-task neural network for solving the problem of recognizing the type of QAM and PSK modulation under parametric a priori uncertainty. Russian Technological Journal. 2023;11(4):49–58. https://doi.org/10.32362/2500-316X-2023-11-4-49-58

15. Abdulkadirov R.I., Alikhanov A.A., Aidamirov N.O., Babayan R.A., Dadalyan A.A., Davydov N.M. Comparative analysis of optimization algorithms using various test functions. In: High-Performance Computing for Solving Applied Problems: Collection of materials of the 12th (69th) Annual Scientific and Practical Conference of Students, Teachers, and Young Scientists of the North Caucasus Federal University. Stavropol; 2025. P. 17–21 (in Russ.). https://elibrary.ru/dqdcfh

16. Gaunov S.R., Baimuradov U.G., Sitnikov S.Yu. Machine learning in Python: using Tensorflow and Scikit-Learn libraries. Ehkonomika i upravlenie: problemy, resheniya = Economics and Management: Problems, Solutions. 2024;8(12-153):72–81 (in Russ.). https://doi.org/10.36871/ek.up.p.r.2024.12.08.009