OPTICAL DEEP LEARNING LANDMINE DETECTION BASED ON LIMITED DATASET OF AERIAL IMAGERY
DOI:
https://doi.org/10.18372/2310-5461.62.18706Keywords:
neural network algorithm, adaptive algorithm, autoregressive process, non-Gaussian disturbancesAbstract
The problem of effective detection of signals remains relevant in many fields of application, including processing of speech signals, analysis of radar prints from moving objects and meteorological formations, and use of information and measurement systems in conditions of radio countermeasures. At the same time, the synthesis of optimal detection procedures is especially complicated under the influence of signals and disturbances, which are characterized by a complex frequency spectrum and cannot be fully described only by Gaussian statistics. Under such conditions, one of the alternatives to the task of non-Gaussianity is the synthesis of robust procedures that can preserve efficiency indicators in a wide class of signal-interference situations. Nevertheless, the growing interest in AI technology has led to the potential use of neural networks to identify signals in complex non-Gaussian environments with uncertainty about probability distributions.
Therefore, the study examines using neural network methodologies for signal detection amidst complex obstacles with prior uncertainty regarding probability distributions. Specifically, it delves into synthesizing and comparing two methods for detecting a harmonic signal with an unknown phase amidst correlated autoregressive noise. The first method entails statistical synthesis of an adaptive detection algorithm, while the second integrates a neural network. It is highlighted that the neural network algorithm excels the statistical algorithm in the action of non-Gaussian impulse interference, yet exhibits reduced efficacy in scenarios featuring Gaussian correlated interference.
References
Ramamurti V., Rao S. S., & Gandhi P. P., “Neural detectors for signals in non-Gaussian noise”. IEEE International Conference on Acoustics Speech and Signal Processing. 1993. doi: 10.1109/icassp.1993.319160.
Allen R., “Automatic earthquake recognition and timing from single traces”. Bull. Seism. Soc. Am, vol. 68, no. 5, pp. 1521–1532. 1978.
Clara E. Yoon et al, “Earthquake detection through computationally efficient similarity search”. Science Advances, Vol. 1, no. 11, e1501057. 2015
Linville L. B., “Global‐and Local‐Scale High‐Resolution Event Catalogs for Algorithm Testing”. Seismological Research Letters, 90(5), pp. 1987–1993. 2019.
Linville L. R., “Global to local high-resolution event catalogs for algorithm testing and source studies”. Seismol. Res. Lett. 2019.
Prokopenko I. G., Dmytruk A. Yu., Prokopenko K. I., “Application of robust algorithms in the problem of detection of moving targets on the background of non-gaussian clutter”, Science-based Technologies, № 1, p. 58–66. 2023. doi: 10.18372/2310-5461.57.17445.
Tian C., Hong M., Li D. and Yuan D., “Deep recurrent neural network for ground-penetrating radar signal denoising”, 2022 4th International Conference on Intelligent Information Processing (IIP), Guangzhou, China, pp. 85–88. 2022. doi: 10.1109/IIP57348.2022.00024.
Geng Z., Yan H., Zhang J. and Zhu D., “Deep-Learning for Radar: A Survey," in IEEE Access, vol. 9, pp. 141800–141818, 2021, doi: 10.1109/ ACCESS.2021.3119561.
Purwins H., Li B., Virtanen T., Schlüter J., Chang S. -Y. and Sainath T., “Deep Learning for Audio
Signal Processing," in IEEE Journal of Selected Topics in Signal Processing, vol. 13, no. 2, pp. 206–219. May 2019. doi: 10.1109/JSTSP. 2019.2908700.
Nanduri A. and Sherry L., “Anomaly detection in aircraft data using Recurrent Neural Networks (RNN), "2016 Integrated Communications Navigation and Surveillance (ICNS), Herndon, VA, USA, pp. 5C2-1-5C2-8. 2016. doi: 10.1109/ ICNSURV.2016.7486356.
Acharya, U. R., Fujita, H., Oh, S. L., Hagiwara, Y., Tan, J. H., & Adam, M. “Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals". Inf. Sci., 415, 190–198. 2017.
Jiang W, Ren Y, Liu Y, Leng J. “Artificial Neural Networks and Deep Learning Techniques Applied to Radar Target Detection: A Review. "Electronics. 11(1), 156. 2022. doi: 10.3390/electronics11010156.
Carrera E. V., Lara F., Ortiz M., Tinoco A. and León R., “Target Detection using Radar Processors based on Machine Learning," 2020 IEEE ANDESCON, Quito, Ecuador, pp. 1–5. 2020. doi: 10.1109/ANDESCON50619.2020.927217.
Prokopenko I., Prokopenko K., Dmytruk A., “Application of Neural Network Technologies in Signal Detection Tasks" IEEE International Conference on Information and Telecommunication Technologies and Radio Electronics (UkrMiCo). Kyiv, Ukraine, pp. 151–156. 2023. doi: 10.1109/ UkrMiCo61577.2023.10380364.9.
Yann LeCun L. B., “Gradient-based learning applied to document recognition”. Proceedings of the IEEE, 86(11), pp. 2278–2324. 1998.
George D. & Huerta E. A., “Deep Neural Networks to Enable Real-time Multimessenger Astrophysics”. Phys. Rev. D 97, 044039. 2018.
Gabbard H., Williams M., Hayes F., & Messenger C., “Matching matched filtering with deep networks in gravitational-wave astronomy”. Phys. Rev. Lett. 120, 141103. 2018.
Wang H., Cao Z., Liu X., Wu S., & Zhu J.-Y, “Gravitational wave signal recognition of O1 data by deep learning”. Phys. Rev. D 101, 104003. 2020.
Krastev P. G., “Real-time detection of gravitational waves from binary neutron stars using artificial neural networks”. Phys. Lett. B 803, 13533. 2020.
López M., Palma I. Di, Drago M., Cerdá-Durán P., & Ricci F., “Deep learning for core-collapse supernova detection“. Phys. Rev. D 103, 063011. 2021.
