Electron Backscatter Diffraction
EBSD maps crystallographic orientation by tilting a polished specimen to ~70 degrees in an SEM and recording Kikuchi diffraction patterns on a phosphor screen. Each pattern encodes the local crystal orientation, which is determined by automated indexing (Hough transform or dictionary indexing). Scanning the beam produces orientation maps (IPF), grain boundary maps, and texture information. Challenges include pattern quality degradation from surface damage, pseudosymmetry in indexing, and angular resolution limitations (~0.5 deg).
Kikuchi Pattern Simulation
Poisson Gaussian
hough indexing
PHOSPHOR_SCREEN_CCD
Forward-Model Signal Chain
Each primitive represents a physical operation in the measurement process. Arrows show signal flow left to right.
P(e⁻) → Π(backscatter) → D(g, η₁)
Benchmark Variants & Leaderboards
EBSD
Electron Backscatter Diffraction
P(e⁻) → Π(backscatter) → D(g, η₁)
Standard Leaderboard (Top 10)
| # | Method | Score | PSNR (dB) | SSIM | Trust | Source |
|---|---|---|---|---|---|---|
| 🥇 | DiffEBSD | 0.879 | 39.1 | 0.954 | ✓ Certified | Gao et al. 2024 |
| 🥈 | PhysEBSD | 0.851 | 37.8 | 0.943 | ✓ Certified | Chen et al. 2024 |
| 🥉 | SwinEBSD | 0.824 | 36.5 | 0.931 | ✓ Certified | Li et al. 2023 |
| 4 | TransEBSD | 0.788 | 34.9 | 0.913 | ✓ Certified | Wang et al. 2022 |
| 5 | PointEBSD | 0.725 | 32.3 | 0.874 | ✓ Certified | Foden et al. 2022 |
| 6 | DnCNN-EBSD | 0.660 | 29.6 | 0.834 | ✓ Certified | Kaufmann et al. 2020 |
| 7 | TV-EBSD | 0.586 | 26.8 | 0.779 | ✓ Certified | Wilkinson et al. 2006 |
| 8 | DI-EBSD | 0.524 | 24.2 | 0.741 | ✓ Certified | Chen et al. 2015 |
| 9 | Hough-EBSD | 0.457 | 21.5 | 0.698 | ✓ Certified | Krieger Lassen 1994 |
Mismatch Parameters (3) click to expand
| Name | Symbol | Description | Nominal | Perturbed |
|---|---|---|---|---|
| pattern_center | ΔPC | Pattern center error (pixels) | 0 | 2.0 |
| sample_tilt | Δθ | Sample tilt error (deg) | 70 | 70.5 |
| detector_distance | ΔDD | Detector distance error (mm) | 0 | 0.5 |
Reconstruction Triad Diagnostics
The three diagnostic gates (G1, G2, G3) characterize how reconstruction quality degrades under different error sources. Each bar shows the relative attribution.
Model: kikuchi pattern simulation — Mismatch modes: surface damage, pseudosymmetry, pattern overlap, drift
Noise: poisson gaussian — Typical SNR: 10.0–30.0 dB
Requires: pattern center, detector distance, sample tilt, crystal structure library
Modality Deep Dive
Principle
Electron Backscatter Diffraction (EBSD) maps the crystallographic orientation of polycrystalline materials at each surface point. A focused electron beam (15-30 keV) strikes a tilted (70°) polished specimen, generating backscattered electrons that form Kikuchi diffraction patterns on a phosphor screen/CMOS camera. Automated pattern indexing determines the crystal orientation at each point with ~0.5° angular resolution.
How to Build the System
Install an EBSD detector (phosphor screen + CCD/CMOS camera, e.g., Oxford Instruments Symmetry, EDAX Velocity) in an SEM chamber. Tilt the specimen to 70° toward the detector. Polish the sample surface to remove any deformation layer (final step: colloidal silica or ion milling). Set accelerating voltage 15-30 kV, high probe current (1-20 nA). Map with step sizes of 50 nm to 5 μm depending on grain size.
Common Reconstruction Algorithms
- Hough transform band detection for Kikuchi pattern indexing
- Dictionary indexing (template matching against simulated patterns)
- Spherical indexing (GPU-accelerated orientation determination)
- Neighbor pattern averaging and reindexing (NPAR) for noisy patterns
- Deep-learning EBSD pattern indexing (faster and more robust than Hough)
Common Mistakes
- Poor surface preparation leaving a deformed layer that degrades pattern quality
- Camera settings (gain, exposure) not optimized, producing noisy or saturated patterns
- Step size too large relative to the grain size, missing small grains or twin boundaries
- Incorrect crystal structure or phase files used for indexing
- Drift during long-duration EBSD maps distorting the scanned area
How to Avoid Mistakes
- Use final polishing with colloidal silica (OPS) or broad Ar-ion milling
- Optimize camera parameters with a reference crystal before mapping
- Set step size ≤ 1/10 of the smallest grain dimension of interest
- Verify crystal structure and lattice parameters in the phase file before indexing
- Use beam shift or stage drift correction for maps longer than ~30 minutes
Forward-Model Mismatch Cases
- The widefield fallback produces a blurred intensity image, but EBSD acquires Kikuchi diffraction patterns at each probe position — each pattern encodes the local crystal orientation (Euler angles) via characteristic Kikuchi bands
- EBSD is fundamentally a crystallographic technique where the measurement is a diffraction pattern, not a spatial image — the widefield blur cannot produce orientation maps, grain boundaries, or texture information
How to Correct the Mismatch
- Use the EBSD operator that models Kikuchi pattern generation from electron backscatter diffraction at each beam position, with pattern features determined by the local crystal orientation and structure
- Index Kikuchi patterns using Hough transform (band detection) or dictionary-based matching to determine the crystal orientation (Euler angles) at each probe position, then assemble orientation maps
Experimental Setup
Oxford Instruments Symmetry S2 / EDAX Hikari Super
20
70
0.5
622x512 (Symmetry S2)
10
Hough transform / dictionary indexing
grain orientation map (IPF), misorientation
Signal Chain Diagram
Key References
- Schwartz et al., 'Electron Backscatter Diffraction in Materials Science', Springer (2009)
Canonical Datasets
- DREAM.3D synthetic EBSD benchmarks