The paper considers a method of generating fractal camouflage structures (grids) using a randomized system of iterative functions. This method allows for changing the base structure (type of mesh), which in turn makes it possible to determine the parameters by which the object can be identified as a fractal camouflage mesh. In the mathematical description of the improved RSIF, the color range parameters (set of colors) are introduced, allowing the fractal structure to be adjusted to the colors of the landscape where the camouflage net will be applied. The choice of colors for the fractal camouflage mesh generator is a critical aspect that affects camouflage effectiveness. Using several shades that correspond to the natural colors of the environment allows for the creation of camouflage structures that are almost impossible to distinguish from real objects on the ground. This approach provides a high degree of concealment and reduces the probability of detecting camouflaged objects even with modern sensor systems. The proposed generation method will enable the formation of an array of information for neural network training. A trained neural network will be able to determine the geometric parameters of the camouflage structure. These parameters can then be used to identify an object hidden under a fractal camouflage structure. The considered generation method allows for the automation of the neural network training process, significantly speeding up the learning process and reducing the need for training data. The proposed approach significantly reduces the risk of human errors and increases the efficiency and effectiveness of military operations. The high accuracy and adaptability of fractal camouflage generated with the help of advanced RSIF and neural networks make this method promising for wide implementation in military technologies.
[1] X. Liang, J. Zhang, L. Zhuo, Y. Li, and Q. Tian, “Small object detection in unmanned aerial vehicle images using feature fusion and scaling-based single shot detector with spatial context analysis,” IEEE Trans. Circuits Syst. Video Technol., vol. 30, no. 6, pp. 1758–1770, 2020. https://doi.org/10.1109/TCSVT.2019.2905881
[2] Y. Liu, C.-Q. Wang, and Y.-J. Zhou, “Camouflaged people detection based on a semi-supervised search identification network,” Def. Technol., vol. 21, pp. 176–183, 2023. https://doi.org/10.1016/j.dt.2021.09.004
[3] S. Wang, D. Zeng, Y. Xu, G. Yang, F. Huang, and L. Chen, “Towards complex scenes: A deep learning-based camouflaged people detection method for snapshot multispectral images,” Def. Technol., vol. 34, pp. 269–281, 2024. https://doi.org/10.1016/j.dt.2023.12.011
[4] Y. Wen, W. Ke, and H. Sheng, “Camouflaged object detection based on deep learning with attention-guided edge detection and multi-scale context fusion,” Appl. Sci. (Basel), vol. 14, no. 6, p. 2494, 2024. https://doi.org/10.3390/app14062494
[5] X. Yang, W.-D. Xu, Q. Jia, and L. Li, “Research on digital camouflage pattern generation algorithm based on adversarial autoencoder network,” Intern. J. Pattern Recognit. Artif. Intell., vol. 34, no. 06, p. 2050017, 2020. https://doi.org/10.1142/S0218001420500172
[6] E. Van der Burg, M. A. Hogervorst, and A. Toet, “Measuring the dynamics of camouflage in natural scenes using convolutional neural networks,” in Target and Background Signatures VIII, 2022. https://doi.org/10.1117/12.2636107
[7] A. Lachkar, T. Gadi, R. Benslimane, L. D’Orazio, and E. Martuscelli, “Textile woven-fabric recognition by using Fourier image-analysis techniques: Part I: A fully automatic approach for crossed-points detection,” J. Text. Inst., vol. 94, no. 3–4, pp. 194–201, 2003. https://doi.org/10.1080/00405000308630608
[8] Van der Burg, E.; Toet, A.; Perone, P.; Hogervorst, M. A. A Convolutional Neural Network as a Potential Tool for Camouflage Assessment. Preprints 2024, 2024012072. https://doi.org/10.20944/preprints202401.2072.v2
[9] Юнак О. М., Климаш М. М., Шпур О. М., Мрак В. Б. Математична модель розпізнавання фрактальних структур з використанням технології нейронних мереж // Infocommunication Technologies and Electronic Engineering = Інфокомунікаційні технології та електронна інженерія. – 2023. – Vol. 3, № 1. – P. 1–9. https://doi.org/10.23939/ictee2023.01.001
[10] Юнак О., Стрихалюк Б., Климаш М. (2023). Ефективність рандомізованою системою ітераційних функцій над детермінованою системою ітераційних функцій при побудові фрактальних зображень з обмеженою роздільною здатністю. Measuring and computing devices in technological processes, (1), 5–12. https://doi.org/10.31891/2219-9365-2023-73-1-1