volume | PIER Journals

Abstract

In the past few years, deep learning has emerged as a transformative force in tackling challenges within the realm of electromagnetic inverse scattering, driving remarkable advances and reshaping conventional approaches. Among them, the physics-assisted learning method that combines traditional inverse scattering algorithms with deep neural networks has demonstrated excellent real-time inversion capability and lower computational complexity. For two-dimensional inverse scattering problems, an approximate solution of the target is first obtained using a linear approximation algorithm, followed by mapping learning from low to high precision with a neural network. To enhance both precision and generalizability, this study integrates a Transformer module into a CBAM U-Net framework, giving rise to a refined architecture aptly named TransAtten U-Net. By retaining certain positional information while enhancing the correlations between features, the overall feature extraction effect is improved. Through simulation experiments, the paper compares the performance of the proposed TransAtten U-Net two-step method, TransAtten U-Net direct method, and CBAM U-Net two-step method. Experimental results demonstrate that the proposed TransAtten U-Net two-step method not only achieves higher accuracy than the other two approaches, but also exhibits a stronger generalization capability across diverse scenarios, along with enabling real-time imaging.

1. Battaglia, Giada Maria, Andrea Francesco Morabito, Roberta Palmeri, and Tommaso Isernia, "Effective non-iterative phase retrieval of 2-D bandlimited signals with applications to antenna characterization and diagnostics," IEEE Transactions on Antennas and Propagation, Vol. 71, No. 8, 6444-6453, 2023.
doi:10.1109/TAP.2023.3283044 Google Scholar

2. Chen, Xudong, Computational Methods for Electromagnetic Inverse Scattering, John Wiley & Sons, 2018.

3. Li, Maokun, Rui Guo, Ke Zhang, Zhichao Lin, Fan Yang, Shenheng Xu, Xudong Chen, Andrea Massa, and Aria Abubakar, "Machine learning in electromagnetics with applications to biomedical imaging: A review," IEEE Antennas and Propagation Magazine, Vol. 63, No. 3, 39-51, 2021.
doi:10.1109/MAP.2020.3043469 Google Scholar

4. Chen, Xudong, Zhun Wei, Li Maokun, and Paolo Rocca, "A review of deep learning approaches for inverse scattering problems (invited review)," Electromagnetic Waves, Vol. 167, 67-81, 2020.
doi:org/10.2528/PIER20030705 Google Scholar

5. Bevacqua, Martina Teresa, Simona Di Meo, Lorenzo Crocco, Tommaso Isernia, and Marco Pasian, "Millimeter-waves breast cancer imaging via inverse scattering techniques," IEEE Journal of Electromagnetics, RF and Microwaves in Medicine and Biology, Vol. 5, No. 3, 246-253, 2021.
doi:10.1109/JERM.2021.3052096 Google Scholar

6. Benny, Ria, Thathamkulam A. Anjit, and Palayyan Mythili, "Deep learning based non-iterative solution to the inverse problem in microwave imaging," Progress In Electromagnetics Research M, Vol. 109, 231-240, 2022.
doi:10.2528/PIERM22010905 Google Scholar

7. Zhou, Huilin, Xin Huang, and Yuhao Wang, "Nonlinear inverse scattering imaging method based on iterative multi-scale network," Chinese Journal of Radio Science, Vol. 37, No. 6, 1019-1024, 2022.
doi:10.12265/j.cjors.2021291 Google Scholar

8. Guo, Rui, Tianyao Huang, Maokun Li, Haiyang Zhang, and Yonina C. Eldar, "Physics-embedded machine learning for electromagnetic data imaging: Examining three types of data-driven imaging methods," IEEE Signal Processing Magazine, Vol. 40, No. 2, 18-31, 2023. Google Scholar

9. Zhou, Yulong, Yu Zhong, Zhun Wei, Tiantian Yin, and Xudong Chen, "An improved deep learning scheme for solving 2-D and 3-D inverse scattering problems," IEEE Transactions on Antennas and Propagation, Vol. 69, No. 5, 2853-2863, 2021.
doi:10.1109/TAP.2020.3027898 Google Scholar

10. Xu, Kuiwen, Zemin Qian, Yu Zhong, Jiangtao Su, Haijun Gao, and Wenjun Li, "Learning-assisted inversion for solving nonlinear inverse scattering problem," IEEE Transactions on Microwave Theory and Techniques, Vol. 71, No. 6, 2384-2395, 2023.
doi:10.1109/TMTT.2022.3228945 Google Scholar

11. Guo, Rui, Zhichao Lin, Tao Shan, Xiaoqian Song, Maokun Li, Fan Yang, Shenheng Xu, and Aria Abubakar, "Physics embedded deep neural network for solving full-wave inverse scattering problems," IEEE Transactions on Antennas and Propagation, Vol. 70, No. 8, 6148-6159, 2022. Google Scholar

12. Yao, He Ming, Wei E. I. Sha, and Lijun Jiang, "Two-step enhanced deep learning approach for electromagnetic inverse scattering problems," IEEE Antennas and Wireless Propagation Letters, Vol. 18, No. 11, 2254-2258, 2019.
doi:10.1109/LAWP.2019.2925578 Google Scholar

13. Gao, Yu, Hongyu Liu, Xianchao Wang, and Kai Zhang, "On an artificial neural network for inverse scattering problems," Journal of Computational Physics, Vol. 448, 110771, 2022.
doi:10.1016/j.jcp.2021.110771 Google Scholar

14. Salucci, Marco, Manuel Arrebola, Tao Shan, and Maokun Li, "Artificial intelligence: New frontiers in real-time inverse scattering and electromagnetic imaging," IEEE Transactions on Antennas and Propagation, Vol. 70, No. 8, 6349-6364, 2022.
doi:10.1109/TAP.2022.3177556 Google Scholar

15. Ronneberger, Olaf, Philipp Fischer, and Thomas Brox, "U-net: Convolutional networks for biomedical image segmentation," Medical Image Computing and Computer-Assisted Intervention --- MICCAI 2015, 234-241, 2015.

16. Xia, Yixin and Siyuan He, "A lightweight deep learning model for full-wave nonlinear inverse scattering problems," Progress In Electromagnetics Research M, Vol. 128, 83-88, 2024.
doi:10.2528/PIERM24071701 Google Scholar

17. Sun, Yu, Zhihao Xia, and Ulugbek S. Kamilov, "Efficient and accurate inversion of multiple scattering with deep learning," Optics Express, Vol. 26, No. 11, 14678-14688, 2018.
doi:10.1364/OE.26.014678 Google Scholar

18. Wei, Zhun and Xudong Chen, "Deep-learning schemes for full-wave nonlinear inverse scattering problems," IEEE Transactions on Geoscience and Remote Sensing, Vol. 57, No. 4, 1849-1860, 2019.
doi:10.1109/TGRS.2018.2869221 Google Scholar

19. Woo, Sanghyun, Jongchan Park, Joon-Young Lee, and In So Kweon, "Cbam: Convolutional block attention module," Proceedings of the European Conference on Computer Vision (ECCV), 3-19, 2018.

20. Vaswani, Ashish, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, and Illia Polosukhin, "Attention is all you need," Advances in Neural Information Processing Systems, Vol. 30, No. 1, 2, 2017. Google Scholar

21. Dosovitskiy, Alexey, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. "An image is worth 16 x 16 words: Transformers for image recognition at scale," ArXiv Preprint ArXiv:2010.11929, 2020. Google Scholar

22. LeCun, Y., C. Cortes, C. Burges, et al. "Mnist handwritten digit database,", 2010. Google Scholar

23. Cakoni, Fioralba, David Colton, and Houssem Haddar, Inverse Scattering Theory and Transmission Eigenvalues, SIAM, 2022.

24. Cohen, Gregory, Saeed Afshar, Jonathan Tapson, and André van Schaik, "EMNIST: Extending MNIST to handwritten letters," 2017 International Joint Conference on Neural Networks (IJCNN), 2921-2926, Anchorage, AK, USA, 2017.
doi:10.1109/IJCNN.2017.7966217

Dr. Zhangyue Zhao

Dr. Chunxia Yang