Public
Eddy Current Imaging — Public Tier
(3 scenes)Full-access development tier with all data visible.
What you get
Measurements (y), ideal forward operator (H), spec ranges, ground truth (x_true), and true mismatch spec.
How to use
Load HDF5 → compare reconstruction vs x_true → check consistency → iterate.
What to submit
Reconstructed signals (x_hat) and corrected spec as HDF5.
Parameter Specifications
✓
True spec visible — use these exact values for Scenario III oracle reconstruction.
| Parameter | Spec Range | True Value | Unit |
|---|---|---|---|
| liftoff_distance | -0.2 – 0.4 | 0.1 | mm |
| conductivity_error | 57.4 – 59.2 | 58.3 | MS/m |
| excitation_frequency_drift | 99.0 – 102.0 | 100.5 | kHz |
| probe_tilt | -0.4 – 0.8 | 0.2 | deg |
InverseNet Baseline Scores
Method: CPU_baseline — Mismatch parameter: nominal
Scenario I (Ideal)
28.49 dB
SSIM 0.8628
Scenario II (Mismatch)
27.85 dB
SSIM 0.8491
Scenario III (Oracle)
30.56 dB
SSIM 0.9105
Per-scene breakdown (4 scenes)
| Scene | PSNR I | SSIM I | PSNR II | SSIM II | PSNR III | SSIM III |
|---|---|---|---|---|---|---|
| scene_00 | 30.40 | 0.8554 | 29.71 | 0.8357 | 31.83 | 0.9028 |
| scene_01 | 24.57 | 0.8899 | 24.78 | 0.8876 | 27.77 | 0.9367 |
| scene_02 | 29.58 | 0.8410 | 27.43 | 0.8472 | 31.43 | 0.9062 |
| scene_03 | 29.41 | 0.8650 | 29.46 | 0.8259 | 31.21 | 0.8962 |
Public Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | DiffEC + gradient | 0.857 | 37.67 | 0.981 | 0.94 | ✓ Certified | Gao et al., NeurIPS 2024 |
| 2 | PhysEC + gradient | 0.823 | 35.93 | 0.973 | 0.87 | ✓ Certified | Chen et al., IEEE Trans. Magn. 2024 |
| 3 | TransEC + gradient | 0.812 | 34.13 | 0.962 | 0.93 | ✓ Certified | Li et al., IEEE Trans. Ind. Electron. 2022 |
| 4 | SwinEC + gradient | 0.809 | 34.5 | 0.964 | 0.89 | ✓ Certified | Wang et al., NDT&E Int. 2023 |
| 5 | ECNN-Defect + gradient | 0.755 | 30.76 | 0.928 | 0.88 | ✓ Certified | Zhang et al., NDT&E Int. 2021 |
| 6 | DnCNN-EC + gradient | 0.701 | 27.29 | 0.865 | 0.91 | ✓ Certified | Gao et al., IEEE Sens. J. 2019 |
| 7 | MUSIC-EC + gradient | 0.652 | 25.43 | 0.815 | 0.86 | ✓ Certified | Skarlatos et al., NDT&E Int. 2012 |
| 8 | TV-EC + gradient | 0.617 | 23.25 | 0.741 | 0.94 | ✓ Certified | Sabbagh et al., IEEE Trans. Magn. 2010 |
| 9 | EC-Deconv + gradient | 0.511 | 19.86 | 0.592 | 0.88 | ✓ Certified | Bowler, J. Appl. Phys. 1994 |
Visible Data Fields
y
H_ideal
spec_ranges
x_true
true_spec
Dataset
Format: HDF5
Scenes: 3
Scoring Formula
0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)
PSNR: 40%
SSIM: 40%
Consistency: 20%