journal article·2023-12

Label-free grain segmentation for optical microscopy images via unsupervised image-to-image translation

Juwon Na, Jaejun Lee, Seong-Hoon Kang, Se-Jong Kim, Seungchul Lee

Publisher page

Abstract

Grain boundaries play an important role in governing the mechanical and physical properties of polycrystalline materials . Therefore, quantitative analysis of grain structure is an important prerequisite for establishing structure-property relationships. However, this quantitative analysis is hampered by the low contrast and non-uniform illumination of optical microscopy images. Although previous studies have used neural networks to detect grain boundaries in optical micrographs, their practical applications have been restricted due to labeling costs for supervised learning and their sensitivity to defects present in a grain structure image. In this paper, we approach grain boundary detection as a real-to-virtual (R2V) translation problem, mapping a single real microstructure to virtual microstructures. With this perspective, our framework benefits from two learning schemes: unsupervised learning and physics-inspired learning. Extensive experiments demonstrate the superiority and generality of our framework. We expect that our approach will facilitate the use of data-driven techniques in the field of quantitative microstructure analysis.
Source: expertdata · Quality: candidate unverified · Source record ↗

Research context

OpenAlex topics: Image Processing Techniques and Applications · Mineral Processing and Grinding · Advanced Electron Microscopy Techniques and Applications

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (51)

Show source-provided references
1. Callister · 2018 · Vol. 9
2. Smith · 1988
3. Strengthening materials by engineering coherent internal boundaries at the nanoscale10.1126/science.1159610 ↗
4. Heterogeneous lamella structure unites ultrafine-grain strength with coarse-grain ductility10.1073/pnas.1517193112 ↗
5. Ultrahigh strength and high electrical conductivity in copper10.1126/science.1092905 ↗
6. Grain boundary engineering enabled high-performance garnet-type electrolyte for lithium dendrite free lithium metal batteries10.1002/smtd.202200667 ↗
7. A review on reducing grain boundaries and morphological improvement of perovskite solar cells from methodology and material-based perspectives10.1002/smtd.201900569 ↗
8. An automated and unbiased grain segmentation method based on directional reflectance microscopy10.1016/j.matchar.2021.110978 ↗
9. The deformation and ageing of mild steel: III discussion of results10.1088/0370-1301/64/9/303 ↗
10. Hall–petch relation and grain boundary slipping in almgsc alloys with fine equiaxed grain structure10.1016/j.matchar.2022.112472 ↗
11. The control of brittleness and development of desirable mechanical properties in polycrystalline systems by grain boundary engineering10.1016/S1359-6454(99)00275-X ↗
12. Goodhew · 2000
13. Fultz · 2012
14. Poulsen · 2004 · vol. 205
15. Synthetic image rendering solves annotation problem in deep learning nanoparticle segmentation10.1002/smtd.202100223 ↗
16. Optical microscopy, encyclopedia of imaging
17. Wojnar · 2019
18. Automatic grain size determination in microstructures using image processing10.1016/j.measurement.2012.06.012 ↗
19. Ceramic image processing using the second curvelet transform and watershed algorithm
20. Algorithm based on regional separation for automatic grain boundary extraction using improved mean shift method10.1088/2051-672X/aab73b ↗
21. Detecting grain boundaries in deformed rocks using a cellular automata approach10.1016/j.cageo.2011.09.008 ↗
22. Method for grain size determination in carbon steels based on the ultimate opening10.1016/j.measurement.2018.09.068 ↗
23. Fast-finecut: grain boundary detection in microscopic images considering 3D information10.1016/j.micron.2018.09.002 ↗
24. Automated grain boundary detection using the level set method10.1016/j.cageo.2008.05.006 ↗
25. New methods for automatic quantification of microstructural features using digital image processing10.1016/j.matdes.2017.12.049 ↗
26. An automated methodology for grain segmentation and grain size measurement from optical micrographs10.1016/j.measurement.2019.03.046 ↗
27. Data augmentation in microscopic images for material data mining
28. Grain boundary detection and second phase segmentation based on multi-task learning and generative adversarial network10.1016/j.measurement.2020.107857 ↗
29. Grain boundary detection based on multi-level loss from feature and adversarial learning10.1109/ACCESS.2020.3011703 ↗
30. Boundary learning by using weighted propagation in convolution network
31. An improved u-net image segmentation method and its application for metallic grain size statistics10.3390/ma15134417 ↗
32. Microstructure representation and reconstruction of heterogeneous materials via deep belief network for computational material design10.1115/1.4036649 ↗
33. Unsupervised machine learning in fractography: evaluation and interpretation10.1016/j.matchar.2021.111551 ↗
34. Unpaired image-to-image translation using cycle-consistent adversarial networks
35. Generative adversarial networks10.1145/3422622 ↗
36. Relative grain boundary energies from triple junction geometry: limitations to assuming the herring condition in nanocrystalline thin films10.1016/j.actamat.2022.118476 ↗
37. Disconnection description of triple-junction motion10.1073/pnas.1820789116 ↗
38. A simple framework for contrastive learning of visual representations
39. Molecular contrastive learning of representations via graph neural networks10.1038/s42256-022-00447-x ↗
40. D. P. Kingma, J. Ba, Adam: A method for stochastic optimization, arXiv preprint arXiv:1412.6980.
41. Training generative adversarial networks with limited data
42. Image processing with imagej
43. Imagej for microscopy
44. 3D: a digital representation environment for the analysis of microstructure in 3D, integrating materials and manufacturing
45. Super-resolving material microstructure image via deep learning for microstructure characterization and mechanical behavior analysis10.1038/s41524-021-00568-8 ↗
46. G. Bradski, The OpenCV library, Dr. Dobb's J. Softw. Tools.
47. The watershed transform: definitions, algorithms and parallelization strategies10.3233/FI-2000-411207 ↗
48. A threshold selection method from gray-level histograms10.1109/TSMC.1979.4310076 ↗
49. Adaptive thresholding using the integral image10.1080/2151237X.2007.10129236 ↗
50. A computational approach to edge detection10.1109/TPAMI.1986.4767851 ↗
51. A deep learning approach for complex microstructure inference10.1038/s41467-021-26565-5 ↗