A machine learning model for textured X-ray scattering and diffraction image denoising
Source abstract
Abstract
Abstract With the advancements in instrumentations of next-generation synchrotron light sources, methodologies for small-angle X-ray scattering (SAXS)/wide-angle X-ray diffraction (WAXD) experiments have dramatically evolved. Such experiments have developed into dynamic and multiscale in situ characterizations, leaving prolonged exposure time as well as radiation-induced damage a serious concern. However, reduction on exposure time or dose may result in noisier images with a lower signal-to-noise ratio, requiring powerful denoising mechanisms for physical information retrieval. Here, we tackle the problem from an algorithmic perspective by proposing a small yet effective machine-learning model for experimental SAXS/WAXD image denoising, allowing more redundancy for exposure time or dose reduction. Compared with classic models developed for natural image scenarios, our model provides a bespoke denoising solution, demonstrating superior performance on highly textured SAXS/WAXD images. The model is versatile and can be applied to denoising in other synchrotron imaging experiments when data volume and image complexity is concerned.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Medical Imaging Techniques and Applications · Advanced X-ray Imaging Techniques · Advanced MRI Techniques and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (48)
Show source-provided references
1. Hura, G. et al. Robust, high-throughput solution structural analyses by small angle X-ray scattering (SAXS). Nat. Methods 6, 606–612 (2009).10.1038/nmeth.1353 ↗
2. Stanić, V. et al. Local structure of human hair spatially resolved by sub-micron X-ray beam. Sci. Rep. 5, 17347 (2015).10.1038/srep17347 ↗
3. Wang, M. et al. SAXS and WAXD study of periodical structure for polyacrylonitrile fiber during coagulation. Polym. Adv. Technol. 26, 136–141 (2014).10.1002/pat.3436 ↗
4. Pauw, B. R. Everything SAXS: small-angle scattering pattern collection and correction. J. Phys. Condens. Matter 25, 383201 (2013).10.1088/0953-8984/25/38/383201 ↗
5. Sui, T. et al. Multiple-length-scale deformation analysis in a thermoplastic polyurethane. Nat. Commun. 6, 6583 (2015).10.1038/ncomms7583 ↗
6. Jeffries, C. M. et al. Small-angle X-ray and neutron scattering. Nat. Rev. Methods Prim. 1, 70 (2021).10.1038/s43586-021-00064-9 ↗
7. Rungswang, W. et al. Time-Resolved SAXS/WAXD under Tensile Deformation: Role of Segmental Ethylene–Propylene Copolymers in Impact-Resistant Polypropylene Copolymers. ACS Applied Polymer Materials 3, 6394–6406 (2021).10.1021/acsapm.1c01159 ↗
8. Hémonnot, C. Y. J. & Köster, S. Imaging of biological materials and cells by X-ray scattering and diffraction. ACS Nano 11, 8542–8559 (2017).10.1021/acsnano.7b03447 ↗
9. Qian, J. et al. Insights into the enhanced reversibility of graphite anode upon fast charging through Li reservoir. ACS Nano 16, 20197–20205 (2022).10.1021/acsnano.2c05428 ↗
10. Schaff, F. et al. Six-dimensional real and reciprocal space small-angle X-ray scattering tomography. Nature 527, 353–356 (2015).10.1038/nature16060 ↗
11. Liebi, M. et al. Nanostructure surveys of macroscopic specimens by small-angle scattering tensor tomography. Nature 527, 349–352 (2015).10.1038/nature16056 ↗
12. Jud, C. et al. X-ray dark-field tomography reveals tooth cracks. Sci. Rep. 11, 14017 (2021).10.1038/s41598-021-93393-4 ↗
13. Georgiadis, M. et al. Nanostructure-specific X-ray tomography reveals myelin levels, integrity and axon orientations in mouse and human nervous tissue. Nat. Commun. 12, 2941 (2021).10.1038/s41467-021-22719-7 ↗
14. Mürer, F. K. et al. Quantifying the hydroxyapatite orientation near the ossification front in a piglet femoral condyle using X-ray diffraction tensor tomography. Sci. Rep. 11, 2144 (2021).10.1038/s41598-020-80615-4 ↗
15. Fratzl, P. Extra dimension for bone analysis. Nature 527, 308–309 (2015).10.1038/527308a ↗
16. Kurihara, H. et al. Elongation induced β- to α-crystalline transformation and microvoid formation in isotactic polypropylene as revealed by time-resolved WAXS/SAXS. Polym. J. 51, 199–209 (2019).10.1038/s41428-018-0141-8 ↗
17. Yang, L., Liu, J., Chodankar, S., Antonelli, S. & DiFabio, J. Scanning structural mapping at the life science X-ray scattering beamline. J. Synchrotron Radiat. 29, 540–548 (2022).10.1107/S1600577521013266 ↗
18. Ilavsky, J. et al. Development of combined microstructure and structure characterization facility for in situ and operando studies at the advanced photon source. J. Appl. Crystallogr. 51, 867–882 (2018).10.1107/S160057671800643X ↗
19. Sarafimov, B. et al. OMNY—a tomography nano crYo stage. Rev. Sci. Instrum. 89, 043706 (2018).10.1063/1.5020247 ↗
20. Ye, D. et al. Preferred crystallographic orientation of cellulose in plant primary cell walls. Nat. Commun. 11, 4720 (2020).10.1038/s41467-020-18449-x ↗
21. Thiyagalingam, J. et al. Scientific machine learning benchmarks. Nat. Rev. Phys. 4, 413–420 (2022).10.1038/s42254-022-00441-7 ↗
22. Zhang, K. et al. Learning deep CNN denoiser prior for image restoration. in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2808–2817 (IEEE, 2017).10.1109/CVPR.2017.300 ↗
23. Tian, C. et al. Image denoising using deep CNN with batch renormalization. Neural Netw. 121, 461–473 (2020).10.1016/j.neunet.2019.08.022 ↗
24. Zhang, K. et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising. IEEE Trans. Image Process. 26, 3142–3155 (2016).10.1109/TIP.2017.2662206 ↗
25. Dabov, K. et al. Image denoising by sparse 3-D transform-domain collaborative filtering. IEEE Trans. Image Process. 16, 2080–2095 (2007).10.1109/TIP.2007.901238 ↗
26. Zhang, K. et al. FFDNet: toward a fast and flexible solution for CNN-based image denoising. IEEE Trans. Image Process. 27, 4608–4622 (2018).10.1109/TIP.2018.2839891 ↗
27. Guo, S. et al. Toward convolutional blind denoising of real photographs. in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 1712–1722 (IEEE, 2019).10.1109/CVPR.2019.00181 ↗
28. Anwar, S. & Barnes, N. Real image denoising with feature attention. in 2019 IEEE/CVF International Conference on Computer Vision (ICCV) 3155–3164 (IEEE, 2019).10.1109/ICCV.2019.00325 ↗
29. Wang, Y. et al. Practical deep raw image denoising on mobile devices. in Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28 (Springer International Publishing, 2020).10.1007/978-3-030-58539-6_1 ↗
30. Hendriksen, A. A. et al. Deep denoising for multi-dimensional synchrotron X-ray tomography without high-quality reference data. Sci. Rep. 11, 11895 (2021).10.1038/s41598-021-91084-8 ↗
31. Bai, T. et al. Deep interactive denoiser (DID) for X-ray computed tomography. IEEE Trans. Med. Imaging 40, 2965–2975 (2021).10.1109/TMI.2021.3101241 ↗
32. Yang, X. et al. Low-dose X-ray tomography through a deep convolutional neural network. Sci. Rep. 8, 2575 (2018).10.1038/s41598-018-19426-7 ↗
33. Chen, H. et al. Low-dose CT with a residual encoder-decoder convolutional neural network. IEEE Trans. Image Process. 36, 2524–2535 (2017).10.1109/TMI.2017.2715284 ↗
34. Shan, H. et al. 3-D convolutional encoder-decoder network for low-dose CT via transfer learning from a 2-D trained network. IEEE Trans. Med. Imag. 37, 1522–1534 (2018).10.1109/TMI.2018.2832217 ↗
35. Niu, Y. et al. Geometrical-based generative adversarial network to enhance digital rock image quality. Phys. Rev. Appl. 15, 064033 (2021).10.1103/PhysRevApplied.15.064033 ↗
36. Bizhani, M., Ardakani, O. H. & Little, E. Reconstructing high fidelity digital rock images using deep convolutional neural networks. Sci. Rep. 12, 4264 (2022).10.1038/s41598-022-08170-8 ↗
37. Yang, Q. et al. Low-dose CT image denoising using a generative adversarial network with Wasserstein distance and perceptual loss. IEEE Trans. Med. Imag. 37, 1348–1357 (2018).10.1109/TMI.2018.2827462 ↗
38. Lee, S. Y. et al. Denoising low-intensity diffraction signals using k-space deep learning: applications to phase recovery. Phys. Rev. Res. 3, 043066 (2021).10.1103/PhysRevResearch.3.043066 ↗
39. Cha, E. et al. Low-dose sparse-view HAADF-STEM-EDX tomography of nanocrystals using unsupervised deep learning. ACS Nano 16, 10314–10326 (2022).10.1021/acsnano.2c00168 ↗
40. Kieffer, J. & Karkoulis, D. PyFAI, a versatile library for azimuthal regrouping. J. Phys. Conf. Ser. 425, 202012 (2013).10.1088/1742-6596/425/20/202012 ↗
41. Ashiotis, G. et al. The fast azimuthal integration Python library: pyFAI. J. Appl. Crystallogr. 48, 510–519 (2015).10.1107/S1600576715004306 ↗
42. Hammersley, A. FIT2D: a multi-purpose data reduction, analysis and visualization program. J. Appl. Crystallogr. 49, 646–652 (2016).10.1107/S1600576716000455 ↗
43. Guo, J. et al. Network decoupling: from regular to depthwise separable convolutions. British Machine Vision Conference (BMVC) (2018).
44. Chollet, F. Xception: deep learning with depthwise separable convolutions. in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 1800–1807 (IEEE, 2017).10.1109/CVPR.2017.195 ↗
45. Paszke, A. et al. Automatic differentiation in PyTorch. NIPS 2017 Workshop on Autodiff (2017).
46. Dong, Y. et al. Exascale image processing for next-generation beamlines in advanced light sources. Nat. Rev. Phys. 4, 427–428 (2022).10.1038/s42254-022-00465-z ↗
47. Li, J. et al. Machine-and-data intelligence for synchrotron science. Nat. Rev. Phys. 3, 766–768 (2021).10.1038/s42254-021-00397-0 ↗
48. Liu, Y. et al. Mamba: a systematic software solution for beamline experiments at HEPS. J. Synchrotron Radiat. 29, 664–669 (2022).10.1107/S1600577522002697 ↗