METHODS OF BUILDING NEURONAL NETWORKS FOR THE IDENTIFICATION OF WEAPONS AND MILITARY EQUIPMENT

Authors
DOI:
https://doi.org/10.34169/2414-0651.2022.2(34).99-107
Keywords:
Array, Array, Array, Array, Array, Array
Abstract

The article substantiates the expediency of using artificial neural networks to identify weapons and military equipment based on the analysis of digital photographs (images) of their constituent elements. It is shown that this problem is caused by an increase in the number of weapons and military equipment samples; requirements for efficiency and the need to automate the process of identification of weapons and military equipment based on the analysis of digital photographs (images). A universal method for constructing an ANN is proposed, which allows using complex neural networks such as AlexNet, GoogleNet, DarkNet-53, DarkNet-19, SgueezeNet, ResNet-50, ShuffleNet, NasNet-Mobile, as well as creating other unique architectures.

It is shown that the problem of identification of WME based on the analysis of digital photographs can be solved using the proposed methodology for constructing an artificial neural networks. An example of the implementation of this technique using the AlexNet artificial neural networks, previously trained on the ImageNet dataset, is given. To solve the problem of increasing the efficiency of WME identification based on the analysis of digital photographs (images) for 3 classes, the initial fully connected layer of the pretrained AlexNet was modified from 1000 to 3 neurons and additional training of the AlexNet ANN was carried out. The effectiveness of the proposed model was tested on a set of 87 images, the total number of classes was 3. Accuracy, learning error were chosen as the main indicators of neural network efficiency. As a result, a new trained model was obtained with an accuracy of identification (classification) of the validation (test) sample – 96 %, which confirms the correct choice of the neural network architecture and training parameters. The use of the proposed technique makes it possible to automate the process of identifying the constituent components of weapons and military equipment based on the analysis of digital photographs (images).

Author Biographies
  1. Vadym Slyusar, Central Scientific Research Institute of Armament and Military Equipment of the Armed Forces of Ukraine

    Doctor of Technical Sciences, Professor

  2. Mykhailo Protsenko, Central Scientific Research Institute of Armament and Military Equipment of the Armed Forces of Ukraine

    Candidate of Technical Science, Senior Researcher

References

Xin, L., Chao, L. & He, L. (2021) Malicious code detection method based on image segmentation and deep residual network RESNET. 2021 Intern. Conf. on Computer Engineering and Application (ICCEA). Pp. 473—480. https://doi.org/10.1109/ICCEA53728.2021.00099. DOI: https://doi.org/10.1109/ICCEA53728.2021.00099

Xie, B., Yang, Z., Yang, L., Luo, R. & Wei, A. (2021) Multi-Scale Fusion With Matching Attention Model: A Novel Decoding Network Cooperated With NAS for Real-Time Semantic Segmentation, in IEEE Transactions on Intelligent Transportation Systems. Pp. 1—11. https://doi.org/10.1109/TITS.2021.3115705. DOI: https://doi.org/10.1109/TITS.2021.3115705

Yang, S., Hou, J., Jia Y., Mei, S. & Du, Q. (2021) Superpixel-Guided Discriminative Low-Rank Representation of Hyperspectral Images for Classification. In IEEE Transactions on Image Processing. Vol. 30. Pp. 8823—8835. https://doi.org/10.1109/TIP.2021.3120675. DOI: https://doi.org/10.1109/TIP.2021.3120675

Slyusar, V., Protsenko, M., Chernukha, A., Gornostal, S., Rudakov, S., Shevchenko, S., Chernikov, O., Kolpachenko, N., Timofeyev, V. & Artiukh, R. (2021). Construction of an advanced method for recognizing monitored objects by a convolutional neural network using a discrete wavelet transform. Eastern-European J. of Enterprise Technologies. No 4(9(112)). Pp. 65—77. https://doi.org/10.15587/1729-4061.2021.238601. DOI: https://doi.org/10.15587/1729-4061.2021.238601

Slyusar, V., Protsenko, M., Chernukha, A., Kovalov, P., Borodych, P., Shevchenko, S., Chernikov, O., Vazhynskyi, S., Bogatov, O. & Khrustalev, K. (2021). Improvement of the model of object recognition in aero photographs using deep convolutional neural networks. Eastern-European J. of Enterprise Technologies. No 5(2(113)). Pp. 6—21. https://doi.org/10.15587/1729-4061.2021.243094. DOI: https://doi.org/10.15587/1729-4061.2021.243094

Slyusar, V., Protsenko, M., Chernukha, A., Melkin, V., Petrova, O., Kravtsov, M., Velma, S., Kosenko, N., Sydorenko, O. & Sobol, M. (2021). Improving a neural network model for semantic segmentation of images of monitored objects in aerial photographs. Eastern-European J. of Enterprise Technologies. No 6(2(114)). Pp. 86—95. https://doi.org/10.15587/1729-4061.2021.248390. DOI: https://doi.org/10.15587/1729-4061.2021.248390

Slyusar, V., Protsenko, M., Chernukha, A., Melkin, V., Biloborodov, O., Samoilenko, M., Kravchenko, O., Kalynychenko, H., Rohovyi, A. & Soloshchuk, M. (2022). Improving the model of object detection on aerial photographs and video in unmanned aerial systems. Eastern-European J. of Enterprise Technologies. No 1(9(115)). Pp. 24—34. https://doi.org/10.15587/1729-4061.2022.252876. DOI: https://doi.org/10.15587/1729-4061.2022.252876

Слюсар В.І., Проценко М.М. Метод розпізнавання зображень об’єктів моніторингу нейронною мережею на основі дискретного вейвлет-перетворення. Prospects and achievements in applied and basic sciences. Матер. IV міжнар. наук. конф. 9−12 лютого 2021 р. Будапешт. C. 600—603. https://doi.org/10.46299/ISG.2021.I.IV. DOI: https://doi.org/10.46299/ISG.2021.I.IV

Слюсар В.І., Проценко М.М. Моніторинг об’єктів нейронною мережею на основі вейвлет-перетворення. Проблеми координації воєнно-технічної та оборонно-промислової політики в Україні. Перспективи розвитку озброєння та військової техніки: зб. матер. IX міжнар. наук.-практ. конф. Київ. 2021. C. 94.

Слюсар В.І., Проценко М.М. Модель детектування об'єктів у відеопотоці з використанням нейронної мережі. IV Міжнар. наук.-практ. конф. «Інтеграція інформаційних систем і інтелектуальних технологій в умовах трансформації інформаційного суспільства», що присвячена 50-ій річниці кафедри інформаційних систем та технологій, 21−22 жовтня 2021 р. Полтава: Полтавський держ. аграрний ун-т. С. 119—122. https://doi.org/10.32782/978-966-289-562-9. DOI: https://doi.org/10.32782/978-966-289-562-9

Zhu, P., Wen, L., Bian, X., Ling, H. & Hu, Q. (2018). Vision meets drones: Achallenge. Pp. 1—11. https://arxiv.org/pdf/1804.07437.pdf.

Офіційний сайт ImageNet. [Електронний ресурс] – Режим доступу: https://www.image-net.org/.

Krizhevsky, A., et al. (2017). ImageNet Classification with Deep Convolutional Neural Networks. Communications of the ACM. No 6 (60). Pp. 84—90. https://doi.org/10.1145/3065386. DOI: https://doi.org/10.1145/3065386

Szegedy, C., Liu, W. & Jia, Y. (2014). Going deeper with convolutions. Pp. 1—12. Available at: https://arxiv.org/pdf/1409.4842v1.pdf.

Yulin, T., Jin, S., Bian, G. & Zhang, Y. (2020). Shipwreck target recognition in side-scan sonar images by improved YOLOv3 model based on transfer learning. IEEE Access 2020. Vol. 8. Pp. 173450—173460. Available at: https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9200467. DOI: https://doi.org/10.1109/ACCESS.2020.3024813

Redmon, J. & Farhadi, A. (2016). YOLO9000: Better, Faster, Stronger. Computer Vision and Pattern Recognition (CVPR). Pp. 1—9. Available at: https://arxiv.org/abs/1612.08242v1.

Офіційний сайт SgueezeNet [Електронний ресурс]. – Режим доступу: https://www.kdnuggets.com/2016/09/deep-learning-reading-group-squeezenet.html.

He, K., Zhang, X., Ren, S. & Sun, J. (2015). Deep Residual Learning for Image Recognition. Computer Vision and Pattern Recognition (CVPR). Pp. 1—12. Available at: https://arxiv.org/pdf/1512.03385.pdf.

Zhang, X., Zhou, X., Lin, M. & Sun, J. (2018) ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Computer Vision and Pattern Recognition (CVPR). Pp. 1—9. Available at: https://arxiv.org/pdf/1707.01083v2.pdf.

Zoph, B., Vasudevan, V., Shlens, J. & Quoc, V, (2018). Learning Transferable Architectures for Scalable Image Recognition Computer Vision and Pattern Recognition (CVPR). Pp. 1—14. Available at: https://arxiv.org/pdf/1707.07012.pdf.

Слюсар В.І. Архітектурно-математичні основи удосконалення нейронних мереж з класифікації зображень. Штучний інтелект. 2022. №1. С. 127 — 138. https://doi.org/10.15407/jai2022.01.127.

Cover Image
Downloads
Published
2022-06-30
Issue
Vol. 34 No. 2 (2022): Weapons and military equipment
Section
AUTOMATED CONTROL SYSTEMS
License

Copyright (c) 2022 Вадим Слюсар ,Михайло Проценко ,Олег Докучаєв

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

METHODS OF BUILDING NEURONAL NETWORKS FOR THE IDENTIFICATION OF WEAPONS AND MILITARY EQUIPMENT. (2022). Weapons and Military Equipment, 34(2), 99-107. https://doi.org/10.34169/2414-0651.2022.2(34).99-107

Most read articles by the same author(s)

1 2 3 > >>