Automated System for Recognizing Air Targets Using Convolutional Neural Networks
Abstract
Purpose. To enhance the efficiency and accuracy of air defense systems through the development and implementation of an automated air target recognition system based on deep learning methods, specifically convolutional neural networks (CNNs).
Method. A quantitative approach was applied using the public Drone Image Classification Dataset containing four image classes (three types of drones and a negative class).
Findings. The developed model demonstrated high classification accuracy of air targets under test conditions, ensuring reliable drone identification and robustness against background noise.
Theoretical implications. The research deepens the scientific understanding of applying convolutional neural networks to real-time air object recognition tasks, contributing to the advancement of computer vision theory in the field of unmanned aerial vehicles.
Practical implications. The obtained results can be utilized for the development and improvement of air defense systems, strategic facility protection, and critical infrastructure security.
Originality of the study. An effective CNN architecture for multi-class air target classification using a public dataset has been proposed, incorporating a negative class to improve system robustness.
Research limitations. The model was tested on static images; to improve its real-world performance, further research should address video stream processing and computational optimization methods for embedded systems.
Downloads
References
Patil, G. U., Kwon, H.-S., Epureanu, B. I., & Popa, B.-I. (2025). Synthetically-trained neural networks for shape classification from measured acoustic scattering. Journal of Sound and Vibration, 618, 119229. https://doi.org/10.1016/j.jsv.2025.119229
Chen, Y., Huang, Y., Xu, C., & Zhao, X. (2024). Multi-fidelity surrogate-based shape optimization using deep neural networks and knowledge-based sampling. Aerospace Science and Technology, 153, 108932. https://doi.org/10.1016/j.ast.2024.108932
Lee, J., Choi, J., & Kim, J. (2025). Deep learning-based real-time structural damage detection using acoustic emission. Engineering Structures, 310, 117006. https://doi.org/10.1016/j.engstruct.2025.117006
Ahmed, S., Wang, X., Li, H., & Zhang, Y. (2024). A hybrid deep learning approach for fault diagnosis of rotating machinery. Measurement, 226, 113161. https://doi.org/10.1016/j.measurement.2024.113161
Wang, Q., Liu, Z., Chen, Z., & Xu, B. (2025). Physics-informed neural networks for solving inverse problems in structural dynamics. Mechanical Systems and Signal Processing, 205, 110807. https://doi.org/10.1016/j.ymssp.2024.110807
Kumar, R., Singh, R., Sharma, M., & Patel, S. (2024). An intelligent fault diagnosis method for roller bearings using ensemble deep learning models. ISA Transactions, 144, 554–564. https://doi.org/10.1016/j.isatra.2024.04.015
Li, J., Zhang, L., Wang, T., & Xu, K. (2024). Damage identification of beam structures using deep autoencoder networks. Structural Control and Health Monitoring, 31(3), e3039. https://doi.org/10.1002/stc.3039
Zhang, H., Gao, Y., Li, P., & Sun, W. (2025). Deep learning for fatigue life prediction of metallic materials. International Journal of Fatigue, 173, 107848. https://doi.org/10.1016/j.ijfatigue.2024.107848
Park, S., Kim, Y., Jung, H., & Lee, D. (2024). A convolutional neural network-based approach for impact force identification in composite structures. Composite Structures, 328, 116962. https://doi.org/10.1016/j.compstruct.2023.116962
Nguyen, T., Pham, H., Tran, D., & Bui, T. (2024). Deep learning-based structural health monitoring using acceleration response. Sensors and Actuators A: Physical, 361, 114529. https://doi.org/10.1016/j.sna.2024.114529
Zhao, H., Wang, Y., Chen, M., & Zhou, J. (2025). Transfer learning for damage detection in structures with limited labeled data. Structural Health Monitoring, 24(2), 123–136. https://doi.org/10.1177/14759217231123609
Liu, Y., Li, H., Zhang, J., & Chen, Y. (2025). Structural damage detection based on deep belief networks and wavelet packet decomposition. Journal of Civil Structural Health Monitoring, 15(1), 1–14. https://doi.org/10.1007/s13349-024-00624-2
Wang, J., Xu, Y., Yang, Z., & Hu, J. (2025). A deep learning framework for real-time damage detection of bridges under moving loads. Computer-Aided Civil and Infrastructure Engineering, 40(5), 482–499. https://doi.org/10.1111/mice.12997
Drone type classification. (n.d.). Kaggle. Available from : https://www.kaggle.com/datasets/balajikartheek/drone-type-classification
Abstract views: 23 PDF Downloads: 9
Copyright (c) 2025 Oleg Lavrov, Andrii Lavrov
This work is licensed under a Creative Commons Attribution 4.0 International License.
The authors agree with the following conditions:
1. Authors retain copyright and grant the journal right of first publication (Download agreement) with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
2. Authors have the right to complete individual additional agreements for the non-exclusive spreading of the journal’s published version of the work (for example, to post work in the electronic repository of the institution or to publish it as part of a monograph), with the reference to the first publication of the work in this journal.
3. Journal’s politics allows and encourages the placement on the Internet (for example, in the repositories of institutions, personal websites, SSRN, ResearchGate, MPRA, SSOAR, etc.) manuscript of the work by the authors, before and during the process of viewing it by this journal, because it can lead to a productive research discussion and positively affect the efficiency and dynamics of citing the published work (see The Effect of Open Access).
