YNR GÖRÜNTÜLERİNDE KARGAŞANIN BASTIRILMASI İÇİN ÖZGÜN TENSÖR GTBA METODU

Abstract

Yere nüfuz eden radar (YNR) görüntülerinde kargaşanın varlığı hedef tespit oranını büyük ölçüde etkilemektedir ve kargaşanın bastırılması için birçok yöntem önerilmiştir. Bu yöntemler temel olarak alçak sıra, alçak sıra ve seyrek ve tensör ayrıştırma yöntemleri olarak gruplanabilir. Temel bileşen analizi (TBA) kargaşa bastırma yöntemler arasında ilk akla gelen yöntemdir ve alçak sıra ailesinde yer alır. Daha sonra, bu yöntem alçak sıra ve seyrek ayrıştırma yöntemi olarak gürbüz temel bileşen analizi (GTBA) adıyla geliştirilmiş ve yüzeye yakın gömülen hedefler ve pürüzlü yüzeyler gibi zorlu durumlarla başa çıkabilir hale gelmiştir. Son zamanlarda önerilen tensör tabanlı yöntemler alçak sıra ve seyrek ayrıştırma problemine alternatif çözümler sağlamaktadır. Bu yöntemlerin sonuçlarından motive olarak, yeni bir ön-dönüşüm adımı ile tensör gürbüz temel bileşen analizi (TGTBA) yöntemi önerilmiştir ve önerilen yöntem TBA ve GTBA yöntemleri ile benzetim veri seti üzerinden karşılaştırılmıştır. Önerdiğimiz yöntem klasik TBA ve yeni önerilen GTBA yöntemlerine karşı hem görsel hem de sayısal olarak üstünlük sağlamıştır.

Keywords

• Yere Nüfuz Eden Radar (YNR)
• Temel Bileşen Analizi (TBA)
• Gürbüz TBA (GTBA)
• Tensör GTBA (TGTBA)
• gprMax

A NOVEL TENSOR RPCA METHOD FOR CLUTTER SUPPRESION IN GPR IMAGES

Abstract

The clutter problem in ground penetrating radar (GPR) images highly effects the peformance of target detection ratio. Various methods have been proposed for clutter suppression purposes in the GPR literature. They can be mainly grouped as low rank, low rank and sparse and tensor-based decomposition methods. Principal component analysis (PCA) and robust principal component analysis (RPCA) are classicle approaches and could be classified as low rank and low rank/sparse decomposition methods, respectively. Recently proposed tensor-based methods provide an alternative perspective of solving the low rank and sparse decomposition to handle challenging situations such as shallowly buried objects or rough surface situations. Motivated by the performance of Tensor-based methods, we propose a new pre-transformation step for tensor robust principal component analysis (TRPCA) and compare it with the PCA and RPCA methods over a simulated GPR dataset. Our proposed method outperforms the classical PCA and recent RPCA methods both visually and quantitatively in terms of clutter removal.

Keywords

• Ground Penetrating Radar (GPR)
• Principal Component Analysis (PCA)
• Robust PCA (RPCA)
• Tensor RPCA (TRPCA)
• gprMax
• gprMax
• gprMax

References

1. Abujarad, F., and Omar, A. (2006). “GPR Data Processing Using the Component-Separation Methods PCA and ICA”. Proceedings of the IEEE International Workshop on Imagining Systems and Techniques (IST 2006), 60-64. doi:10.1109/IST.2006.1650776.
2. Candès, E. J., Li, X., Ma, Y., and Wright, J. (2011). “Robust Principal Component Analysis?”. Journal of the ACM, 58(3), 1-37. doi:10.1145/1970392.1970395.
3. Daniels, D. J. (2005). “Ground Penetrating Radar”. Encyclopedia of RF and Microwave Engineering, John Wiley & Sons.
4. Kumlu, D., and Erer, I. (2018). “The Multiscale Directional Neighborhood Filter and its Application to Clutter Removal in GPR Data”. Signal, Image and Video Processing (SIViP), 12(7), 1237-1244.
5. Kumlu, D., and Erer, I. (2019a). “Clutter Removal Techniques in Ground Penetrating Radar for Landmine Detection: A Survey”. In H. Tozan and M. Karatas (Eds.), Operations Research for Military Organizations (pp. 375- 399). Hershey, PA: IGI Global.
6. Kumlu, D., and Erer, I. (2019b), “Improved Cutter Removal in GPR by Robust Nonnegative Matrix Factorization”. In IEEE Geoscience and Remote Sensing Letters, 17(6), 958-962. Retrieved from https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8836507
7. Kumlu, D., and Erer, I. (2020a). “GPR Clutter Reduction by Robust Orthonormal Subspace Learning”. In IEEE Access, Vol. 8, 74145-74156. Retrieved from https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9069272
8. Kumlu, D., Erer, I., and Kaplan, N.H., (2020b). “Low Complexity Clutter Removal in GPR Images via Lattice Filters”, Digital Signal Processing, Vol. 101, 102724. doi:10.1016/j.dsp.2020.102724.

9. Lu, C., Feng, J., Chen, Y., Liu, W., Lin, Z., and Yan, S. (2019). “Tensor Robust Principal Component Analysis with a New Tensor Nuclear Norm”. IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(4), 925-938. doi:10.1109/tpami.2019.2891760.
10. Song, X., Xiang, D., Zhou, K., and Su, Y. (2017). “Improving RPCA-based Clutter Suppression in GPR Detection of Antipersonnel Mines”, IEEE Geoscience and Remote Sensing Letters, 14(8), 1338-1342. Retrieved from https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7970124
11. Song, X., Liu, T., Xiang, D., and Su, Y. (2019). “GPR Antipersonnel Mine Detection Based on Tensor Robust Principal Analysis”. Remote Sensing, 11(8), 984.doi:10.3390/rs11080984.
12. Tivive, F. H. C., Bouzerdoum, A., and Abeynayake, C. (2019). “GPR Target Detection by Joint Sparse and Low-Rank Matrix Decomposition”. IEEE Transactions on Geoscience and Remote Sensing, 57(5), 2583-2595. Retrieved from https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8522060
13. Verma, P. K., Gaikwad, A. N., Singh, D., and Nigam, M. J. (2009). “Analysis of Clutter Reduction Techniques for through Wall Imaging in UWB Range”. Progress In Electromagnetics Research, Vol. 17, 29-48. Retrieved from http://www.jpier.org/PIERB/pierb17/03.09060903.pdf
14. Warren, C., Giannopoulos, A., and Giannakis, I. (2016). “gprMax: Open Source Software to Simulate Electromagnetic Wave Propagation for Ground Penetrating Radar”. Computer Physics Communications, Vol. 209, 163-170. doi:10.1016/j.cpc.2016.08.020.
15. Wold, S., Esbensen, K., and Geladi, P. (1987). “Principal Component Analysis”. Chemometrics and Intelligent Laboratory Systems, 2(1-3), 37-52. doi:10.1016/0169-7439(87)80084-9.