Human perception-inspired grain segmentation refinement using conditional random fields
Source abstract
Abstract
Automated detection of grain boundaries in electron microscope images of polycrystalline materials could help accelerate the nanoscale characterization of myriad engineering materials and novel materials under scientific research. Accurate segmentation of interconnected line networks, such as grain boundaries in polycrystalline material microstructures, poses a significant challenge due to the fragmented masks produced by conventional computer vision algorithms, including convolutional neural networks. These algorithms struggle with thin masks, often necessitating post-processing for effective contour closure and continuity. Previous approaches in this domain have typically relied on custom post-processing techniques that are problem-specific and heavily dependent on the quality of the mask obtained from a computer vision algorithm. Addressing this issue, this paper introduces a fast, high-fidelity post-processing technique that is universally applicable to segmentation masks of interconnected line networks. Leveraging domain knowledge about grain boundary connectivity, this method employs conditional random fields and perceptual grouping rules to refine segmentation masks of any image with a discernible grain structure. This approach significantly enhances segmentation mask accuracy by correctly reconstructing fragmented grain boundaries in electron microscopy images of a polycrystalline oxide. The refinement improves the statistical representation of the microstructure, reflected by a 51 % improvement in a grain alignment metric that provides a more physically meaningful assessment of complex microstructures than conventional metrics. This method enables rapid and accurate characterization, facilitating an unprecedented level of data analysis and improving the understanding of grain boundary networks, making it suitable for a range of disciplines where precise segmentation of interconnected line networks is essential.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Image Processing and 3D Reconstruction · Image and Object Detection Techniques · 3D Shape Modeling and Analysis
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (83)
Show source-provided references
1. The role of grain boundary structure and crystal orientation on crack growth asymmetry in aluminum10.1016/j.msea.2014.08.083 ↗
2. Enhanced radiation damage tolerance of amorphous interphase and grain boundary complexions in cu-ta10.1007/s11837-024-06382-z ↗
3. An embedded-atom method potential parameterized for sulfur-induced embrittlement of nickel10.1088/1361-651X/ab4c48 ↗
4. Spectrum of embrittling potencies and relation to properties of symmetric-tilt grain boundaries10.1016/j.actamat.2020.116527 ↗
5. A machine learning framework for the prediction of grain boundary segregation in chemically complex environments10.1088/1361-651X/ad585f ↗
6. Chemical order transitions within extended interfacial segregation zones in NbMoTaW10.1063/5.0122502 ↗
7. Aksoy · 2023
8. Conditional random fields meet deep neural networks for semantic segmentation: combining probabilistic graphical models with deep learning for structured prediction10.1109/MSP.2017.2762355 ↗
9. RoadTracer: automatic extraction of road networks from aerial images
10. Linking macroscopic and Nanoscopic ionic conductivity: a Semiempirical framework for characterizing grain boundary conductivity in polycrystalline ceramics10.1021/acsami.9b15933 ↗
11. Enhanced ionic conductivity in electroceramics by nanoscale enrichment of grain boundaries with high solute concentration10.1039/C7NR06941C ↗
12. The role of doping and microstructure on hydrogen solubility in monoclinic ZrO2: experimental validations of simulated defect chemistry10.1016/j.actamat.2020.04.020 ↗
13. Electrical conductivity and grain boundary composition of Gd-doped and Gd/Pr co-doped ceria10.1016/j.ssi.2014.12.006 ↗
14. Recrystallization and grain growth10.1016/0502-8205(52)90009-9 ↗
15. Grain boundary complexion transitions10.1146/annurev-matsci-081619-114055 ↗
16. Geodesic active contours10.1023/A:1007979827043 ↗
17. Deep learning for cardiac image segmentation: a review10.3389/fcvm.2020.00025 ↗
18. Chen · 2017
19. Cheng · 2021
20. Hierarchical exploration of volumes using multilevel segmentation of the intensity-gradient histograms10.1109/TVCG.2012.231 ↗
21. The use of image analysis for sintering investigations: the example of CeO2 doped with TiO210.1016/j.jeurceramsoc.2004.09.003 ↗
22. Fast approximate energy minimization with label costs
23. Grain boundary detection in microstructure images using computational intelligence10.1016/j.compind.2005.05.012 ↗
24. A primer on selecting grain boundary sets for comparison of interfacial fracture properties in molecular dynamics simulations10.1038/s41598-017-08637-z ↗
25. Ensuring a connected structure for retinal vessels deep-learning segmentation
26. Segregation competition and complexion coexistence within a polycrystalline grain boundary network10.1016/j.actamat.2021.117213 ↗
27. Deep convolutional neural networks as a rapid screening tool for complex additively manufactured structures
28. Image segmentation and analysis via multiscale gradient watershed hierarchies10.1109/83.736688 ↗
29. Detecting grain boundaries in deformed rocks using a cellular automata approach10.1016/j.cageo.2011.09.008 ↗
30. He
31. He · 2015
32. The sofa problem10.1093/comjnl/11.3.299 ↗
33. A segmentation scheme for complex neuronal Arbors and application to vibration sensitive neurons in the honeybee brain10.3389/fninf.2018.00061 ↗
34. Jahangard · 2020
35. Musculoskeletal MR image segmentation with artificial intelligence10.1016/j.yacr.2022.04.010 ↗
36. Kingma · 2017
37. Compositionally complex perovskite oxides: discovering a new class of solid electrolytes with interface-enabled conductivity improvements10.1016/j.matt.2023.05.035 ↗
38. Krähenbühl · 2012
39. Microstructure classification in the unsupervised context10.1016/j.actamat.2021.117434 ↗
40. Segmentation and classification approaches of clinically relevant curvilinear structures: a review10.1007/s10916-023-01927-2 ↗
41. Lymphatic vessel segmentation in optical coherence tomography by adding U-net-based CNN for artifact minimization10.1364/BOE.389373 ↗
42. Grain boundary detection and second phase segmentation based on multi-task learning and generative adversarial network10.1016/j.measurement.2020.107857 ↗
43. Quadruple nodes and grain boundary connectivity in three dimensions10.1016/j.actamat.2013.10.033 ↗
44. Image quality assessment based on gradient similarity10.1109/TIP.2011.2175935 ↗
45. Monte Carlo simulation of order-disorder transition in refractory high entropy alloys: a data-driven approach10.1016/j.commatsci.2020.110135 ↗
46. Fully convolutional networks for semantic segmentation
47. Learning to detect natural image boundaries using local brightness, color, and texture cues10.1109/TPAMI.2004.1273918 ↗
48. Two-dimensional motion of idealized grain boundaries10.1063/1.1722511 ↗
49. Vapour phase growth and grain boundary structure of molybdenum disulphide atomic layers10.1038/nmat3673 ↗
50. Edge and line oriented contour detection: state of the art10.1016/j.imavis.2010.08.009 ↗
51. Improved representations of misorientation information for grain boundary science and engineering10.1016/j.pmatsci.2012.04.002 ↗
52. Automated grain boundary detection for bright-field transmission electron microscopy images via U-net10.1093/micmic/ozad115 ↗
53. Z-contrast stem for materials science10.1016/0304-3991(89)90173-3 ↗
54. Automatic grain size determination in microstructures using image processing10.1016/j.measurement.2012.06.012 ↗
55. Optimized and autonomous machine learning framework for characterizing pores, particles, grains and grain boundaries in microstructural images10.1016/j.commatsci.2021.110524 ↗
56. Diffraction contrast STEM of dislocations: imaging and simulations10.1016/j.ultramic.2011.07.001 ↗
57. Differential evolution
58. Deep learning for mapping element distribution of high-entropy alloys in scanning transmission electron microscopy images10.1016/j.commatsci.2021.110905 ↗
59. Rezanejad · 2021
60. Ronneberger · 2015
61. Automated grain boundary detection by CASRG10.1016/j.jsg.2005.12.010 ↗
62. Image segmentation based on watershed and edge detection techniques
63. Impact of the segregation energy spectrum on the enthalpy and entropy of segregation10.1016/j.actamat.2021.117393 ↗
64. A high-throughput technique for determining grain boundary character non-destructively in microstructures with through-thickness grains10.1038/npjcompumats.2016.16 ↗
65. TextonBoost for image understanding: multi-class object recognition and segmentation by jointly Modeling texture, layout, and context10.1007/s11263-007-0109-1 ↗
66. Simonyan · 2015
67. An introduction to conditional random fields10.1561/2200000013 ↗
68. Heterointerface and grain boundary energies, and their influence on microstructure in multiphase ceramics10.1016/j.actamat.2022.117685 ↗
69. Correlations of grain boundary segregation to sintering techniques in a three-phase ceramic10.1016/j.mtla.2020.100890 ↗
70. New data-driven interacting-defect model describing Nanoscopic grain boundary compositions in ceramics10.1021/acs.jpcc.0c05713 ↗
71. Applications of deep learning in electron microscopy10.1093/jmicro/dfab043 ↗
72. A review of grain boundary and Heterointerface characterization in polycrystalline oxides by (scanning) transmission electron microscopy10.3390/cryst11080878 ↗
73. Grain segmentation in atomistic simulations using orientation-based iterative self-organizing data analysis10.1016/j.mtla.2022.101314 ↗
74. Viewpoint: can symmetric tilt grain boundaries represent polycrystals?10.1016/j.scriptamat.2023.115716 ↗
75. Probing multiscale disorder in Pyrochlore and related complex oxides in the transmission electron microscope: a review10.3389/fchem.2021.743025 ↗
76. Stochastic completion fields: a neural model of illusory contour shape and salience10.1162/neco.1997.9.4.837 ↗
77. Exact algorithms for NP-hard problems: A survey
78. Anisotropy of segregation at grain boundaries and surfaces10.1007/BF02586096 ↗
79. Connected contours: A new contour completion model that respects the closure effect
80. A petascale automated imaging pipeline for mapping neuronal circuits with high-throughput transmission electron microscopy10.1038/s41467-020-18659-3 ↗
81. Unified approach to pavement crack and sealed crack detection using Preclassification based on transfer learning10.1061/(ASCE)CP.1943-5487.0000736 ↗
82. A neural network approach to segment brain blood vessels in digital subtraction angiography10.1016/j.cmpb.2019.105159 ↗
83. Conditional random fields as recurrent neural networks