Computer Vision for UAV-based Reconnaissance under Conditions of Modern Warfare

Authors

DOI:

https://doi.org/10.18372/1990-5548.86.20560

Keywords:

unmanned aerial vehicle, computer vision, deep learning, object detection, military reconnaissance, neural networks

Abstract

This paper considers the application of computer vision and deep learning methods for automated aerial reconnaissance using unmanned aerial vehicles under the conditions of modern warfare. The main classes of reconnaissance objects are analyzed, including military vehicles, fortifications, artillery positions, and groups of personnel. An approach to building an object detection system based on deep neural networks is proposed, in particular using YOLO-type detectors and U-Net segmentation models. The process of data preparation and augmentation with consideration of combat factors (smoke, explosions, low illumination, image shift, and noise) is described. An experimental evaluation of object detection quality under different scenarios is performed. It is shown that the use of specially adapted augmentation significantly increases the robustness of the models to interference. The limitations of the proposed approach and directions for further research are discussed.

Author Biography

Anatoly Kot , National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute,”

PhD Student

Artificial Intelligence Department

Educational and Research Institute for Applied System Analysis

References

J. Redmon and A. Farhadi, “YOLOv3: An Incremental Improvement,” arXiv preprint arXiv:1804.02767, 2018. https://arxiv.org/abs/1804.02767

A. Bochkovskiy, C.-Y. Wang, and H.-Y. M. Liao, “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv preprint arXiv:2004.10934, 2020. https://arxiv.org/abs/2004.10934

O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” Proc. MICCAI, 2015. https://arxiv.org/abs/1505.04597 https://doi.org/10.1007/978-3-319-24574-4_28

T.-Y. Lin, P. Goyal, R. Girshick, K. He, and P. Dollár, “Focal Loss for Dense Object Detection,” IEEE International Conference on Computer Vision (ICCV), 2017. https://arxiv.org/abs/1708.02002 https://doi.org/10.1109/ICCV.2017.324

M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “The PASCAL Visual Object Classes (VOC) Challenge,” International Journal of Computer Vision, vol. 88, no. 2, pp. 303–338, 2010. https://link.springer.com/article/10.1007/s11263-009-0275-4

G.-S. Xia, J. Hu, F. Hu, X. Shi, X. Bai, Y. Zhong, L. Zhang, and X. Lu, “DOTA: A Large-Scale Dataset for Object Detection in Aerial Images,” Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. https://arxiv.org/abs/1711.10398 https://doi.org/10.1109/CVPR.2018.00418

D. Lam, et al., “xView: Objects in Context in Overhead Imagery,” Proc. IEEE CVPR, 2018. https://arxiv.org/abs/1802.07856

S. Shah, D. Dey, C. Lovett, and A. Kapoor, “AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles,” Proc. IEEE CVPR, 2017. https://arxiv.org/abs/1705.05065 https://doi.org/10.1007/978-3-319-67361-5_40

A. G. Howard, et al., “MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications,” arXiv preprint arXiv:1704.04861, 2017. https://arxiv.org/abs/1704.04861

A. Garcia, A. Gordon, J. Hurst, S. Konecny, S. Kim, and K. An, “Glider: A Deep Learning Framework for Edge Inference,” arXiv preprint arXiv:2003.03119, 2020. https://arxiv.org/abs/2003.03119

C. Shorten and T. M. Khoshgoftaar, “A Survey on Image Data Augmentation for Deep Learning,” Journal of Big Data, vol. 6, no. https://journalofbigdata.springeropen.com/articles/10.1186/s40537-019-0197-0 60, 2019.

I. Goodfellow, Y. Bengio, and A. Courville, “Deep Learning,” MIT Press, Cambridge, 2016. https://www.deeplearningbook.org

Downloads

Published

2025-12-13

Issue

Vol. 4 No. 86 (2025)

Section

COMPUTER SCIENCES AND INFORMATION TECHNOLOGIES

License

Authors who publish with this journal agree to the following terms:

Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.

Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.

Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).