Nuclear Engineering and Technology

Korean Nuclear Society (한국원자력학회)
권호 표지
DOI QR코드

DOI QR Code

A high-density gamma white spots-Gaussian mixture noise removal method for neutron images denoising based on Swin Transformer UNet and Monte Carlo calculation

  • Di Zhang (Key Laboratory of Neutronics and Radiation Safety, Hefei Institutes of Physical Science, Chinese Academy of Sciences) ;
  • Guomin Sun (Key Laboratory of Neutronics and Radiation Safety, Hefei Institutes of Physical Science, Chinese Academy of Sciences) ;
  • Zihui Yang (Key Laboratory of Neutronics and Radiation Safety, Hefei Institutes of Physical Science, Chinese Academy of Sciences) ;
  • Jie Yu (Key Laboratory of Neutronics and Radiation Safety, Hefei Institutes of Physical Science, Chinese Academy of Sciences)
  • Received : 2022.11.18
  • Accepted : 2023.11.06
  • Published : 2024.02.25
Download PDF
⟨ Previous Next ⟩

Abstract

During fast neutron imaging, besides the dark current noise and readout noise of the CCD camera, the main noise in fast neutron imaging comes from high-energy gamma rays generated by neutron nuclear reactions in and around the experimental setup. These high-energy gamma rays result in the presence of high-density gamma white spots (GWS) in the fast neutron image. Due to the microscopic quantum characteristics of the neutron beam itself and environmental scattering effects, fast neutron images typically exhibit a mixture of Gaussian noise. Existing denoising methods in neutron images are difficult to handle when dealing with a mixture of GWS and Gaussian noise. Herein we put forward a deep learning approach based on the Swin Transformer UNet (SUNet) model to remove high-density GWS-Gaussian mixture noise from fast neutron images. The improved denoising model utilizes a customized loss function for training, which combines perceptual loss and mean squared error loss to avoid grid-like artifacts caused by using a single perceptual loss. To address the high cost of acquiring real fast neutron images, this study introduces Monte Carlo method to simulate noise data with GWS characteristics by computing the interaction between gamma rays and sensors based on the principle of GWS generation. Ultimately, the experimental scenarios involving simulated neutron noise images and real fast neutron images demonstrate that the proposed method not only improves the quality and signal-to-noise ratio of fast neutron images but also preserves the details of the original images during denoising.

Keywords

References

  1. X. Yuan, S. Han, Single-pixel neutron imaging with artificial intelligence: breaking the barrier in multi-parameter imaging, sensitivity, and spatial resolution, Innovation 2 (2) (2021), https://doi.org/10.1016/j.xinn.2021.100100.
  2. R. Zboray, R. Adams, Z. Kis, Fast neutron radiography and tomography at a 10 mw research reactor beamline, Appl. Radiat. Isot. 119 (2017) 43-50, https://doi.org/10.1016/j.apradiso.2016.10.012.
  3. S. Cho, J. Kim, T. Kim, H. Sato, I. Huh, N. Cho, Neutron imaging for metallurgical characteristics of iron products manufactured with ancient Korean iron making techniques, Nucl. Eng. Technol. 53 (5) (2021) 1619-1625, https://doi.org/10.1016/j.net.2020.11.007.
  4. E.H. Lehmann, D. Mannes, M. Strobl, B. Walfort, A. Losko, B. Schillinger, M. Schulz, S.C. Vogel, D. Schaper, D.C. Gautier, et al., Improvement in the spatial resolution for imaging with fast neutrons, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 988 (2021), 164809, https://doi.org/10.1016/j.nima.2020.164809.
  5. A. Buades, B. Coll, J.-M. Morel, A non-local algorithm for image denoising, in: 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05), vol. 2, Ieee, 2005, pp. 60-65, https://doi.org/10.1109/CVPR.2005.38.
  6. K. Dabov, A. Foi, V. Katkovnik, K. Egiazarian, Image denoising by sparse 3-d transform-domain collaborative filtering, IEEE Trans. Image Process. 16 (8) (2007) 2080-2095, https://doi.org/10.1109/TIP.2007.901238.
  7. H. Ye, S. Qiao, C. Zhao, Improved pm method for neutron image denoising, in: 2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), IEEE, 2019, pp. 1-6, https://doi.org/10.1109/CISP-BMEI48845.2019.8965744.
  8. C. Zhao, Y. Yan, Y. Wang, S. Qiao, White spots noise removal of neutron images using improved robust principal component analysis, Fusion Eng. Des. 156 (2020), 111739, https://doi.org/10.1016/j.fusengdes.2020.111739.
  9. A. Hindasyah, D. Sudiana, D. Gunawan, A novel technique for removal of high density white spot noise from digital neutron radiographic images, At. Indones. 44 (3) (2018) 165-171, https://doi.org/10.17146/aij.2018.846.
  10. D.A. Hindasyah, D. Sudiana, D. Gunawan, A simple filtering algorithm for gamma spot removal in neutron radiography, Int. J. Adv. Comput. Sci. Technol. 4 (2) (2015) 1-5. http://warse.org/pdfs/2015/iccsit2015sp01.pdf.
  11. H. Li, B. Schillinger, E. Calzada, L. Yinong, M. Muehlbauer, An adaptive algorithm for gamma spots removal in ccd-based neutron radiography and tomography, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 564 (1) (2006) 405-413, https://doi.org/10.1016/j.nima.2006.04.063.
  12. K. Osterloh, T. Bucherl, C.L. Von Gostomski, U. Zscherpel, U. Ewert, S. Bock, Filtering algorithm for dotted interferences, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 651 (1) (2011) 171-174, https://doi.org/10.1016/j.nima.2011.01.107.
  13. L. He, S.D. Miller, H. Qi, Advanced processing methodologies improve neutron radiograph image quality, in: ASNT 22nd Research Symposium 2013, 2013, pp. 28-32.
  14. Q. Shuang, Z. Wei-jing, S. Jia-ning, Neutron radiographic image restoration using bm3d frames and nonlinear variance stabilization, Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip. 789 (2015) 95-100, https://doi.org/10.1016/j.nima.2015.04.005.
  15. C. Zhao, Y. Yan, H. Li, T. Zhang, S. Qiao, An effective gamma white spots removal method for ccd-based neutron images denoising, Fusion Eng. Des. 150 (2020), 111375, https://doi.org/10.1016/j.fusengdes.2019.111375.
  16. J. Yang, C. Zhao, S. Qiao, et al., Deep learning methods for neutron image restoration, Ann. Nucl. Energy 188 (2023), 109820.
  17. H. Aoki, Y. Liu, T. Yamashita, Deep learning approach for an interface structure analysis with a large statistical noise in neutron reflectometry, Sci. Rep. 11 (1) (2021), 22711.
  18. C.-M. Fan, T.-J. Liu, K.-H. Liu, Sunet: Swin Transformer Unet for Image Denoising, 2022, https://doi.org/10.48550/arXiv.2202.14009 arXiv preprint arXiv:2202.14009.
  19. Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, B. Guo, Swin transformer: hierarchical vision transformer using shifted windows, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, https://doi.org/10.48550/arXiv.2103.14030, 10 012-10 022.
  20. K.-Y. Wen, T.-J. Liu, K.-H. Liu, D.-Y. Chao, Identifying poultry farms from satellite images with residual dense u-net, in: 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), IEEE, 2020, pp. 102-107, https://doi.org/10.1109/SMC42975.2020.9283340.
  21. H. Cao, Y. Wang, J. Chen, D. Jiang, X. Zhang, Q. Tian, M. Wang, Swin-unet: Unetlike Pure Transformer for Medical Image Segmentation, 2021, https://doi.org/10.48550/arXiv.2105.05537 arXiv preprint arXiv:2105.05537.
  22. K.-C. Chang, T.-J. Liu, K.-H. Liu, D.-Y. Chao, Locating waterfowl farms from satellite images with parallel residual unet architecture, in: 2020 IEEE International Conference on Systems, Man, and Cybernetics, (SMC). IEEE, 2020, pp. 114-119, https://doi.org/10.1109/SMC42975.2020.9283182.
  23. Y.-Z. Su, T.-J. Liu, K.-H. Liu, H.-H. Liu, S.-C. Pei, Image inpainting for random areas using dense context features, in: 2019 IEEE International Conference on Image Processing (ICIP), IEEE, 2019, pp. 4679-4683, https://doi.org/10.1109/ICIP.2019.8803450.
  24. J. Liang, J. Cao, G. Sun, K. Zhang, L. Van Gool, R. Timofte, Swinir: image restoration using swin transformer, in: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021, pp. 1833-1844, https://doi.org/10.48550/arXiv.2108.10257.
  25. C. Miao, L. Xie, F. Wan, et al., A large-scale security inspection X-ray benchmark for prohibited item discovery in overlapping images, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2019, pp. 15-20. Long Beach, CA, USA.
  26. R. Komatsu, H. Woo, Y. Tamura, A. Yamashita, H. Asama, Gamma-ray image noise generation using energy-image converter based on image histogram, in: 2021 IEEE/SICE International Symposium on System Integration (SII), IEEE, 2021.
  27. K. Simonyan, A. Zisserman, Very Deep Convolutional Networks for Large-Scale Image Recognition, 2014, https://doi.org/10.48550/arXiv.1409.1556 arXiv preprint arXiv:1409.1556.
  28. Z. Wang, A.C. Bovik, H.R. Sheikh, E.P. Simoncelli, Image quality assessment: from error visibility to structural similarity, IEEE Trans. Image Process. 13 (4) (2004) 600-612, https://doi.org/10.1109/TIP.2003.819861.
  29. M.V. Afonso, J.M.R. Sanches, Blind inpainting using 0 and total variation regularization, IEEE Trans. Image Process. 24 (7) (2015) 2239-2253, https://doi.org/10.1109/TIP.2015.2417505.
  30. K. Zhang, W. Zuo, Y. Chen, D. Meng, L. Zhang, Beyond a Gaussian denoiser: residual learning of deep cnn for image denoising, IEEE Trans. Image Process. 26 (7) (2017) 3142-3155, https://doi.org/10.1109/TIP.2017.2662206.
  31. K. Zhang, W. Zuo, L. Zhang, FFDNet: toward a fast and flexible solution for CNN based image denoising, IEEE Trans. Image Process. (2017), https://doi.org/10.1109/TIP.2018.2839891, 1-1.
  32. B. Park, S. Yu, J. Jeong, Densely connected hierarchical network for image denoising, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019, 0-0.