Public
Near-field Scanning Optical Microscopy (NSOM) — 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 |
|---|---|---|---|
| tip_sample_distance | 2.0 – 26.0 | 14.0 | nm |
| aperture_size_error | -4.0 – 8.0 | 2.0 | - |
| topographic_coupling | -6.0 – 12.0 | 3.0 | - |
| far_field_background | -4.0 – 8.0 | 2.0 | - |
InverseNet Baseline Scores
Method: CPU_baseline — Mismatch parameter: nominal
Scenario I (Ideal)
16.81 dB
SSIM 0.5697
Scenario II (Mismatch)
18.59 dB
SSIM 0.4867
Scenario III (Oracle)
19.16 dB
SSIM 0.6312
Per-scene breakdown (4 scenes)
| Scene | PSNR I | SSIM I | PSNR II | SSIM II | PSNR III | SSIM III |
|---|---|---|---|---|---|---|
| scene_00 | 15.43 | 0.5769 | 18.62 | 0.5430 | 17.94 | 0.6537 |
| scene_01 | 18.09 | 0.5684 | 19.05 | 0.4624 | 19.61 | 0.6069 |
| scene_02 | 18.65 | 0.5692 | 18.97 | 0.4479 | 19.89 | 0.6119 |
| scene_03 | 15.06 | 0.5643 | 17.74 | 0.4936 | 19.22 | 0.6522 |
Public Tier Leaderboard
| # | Method | Score | PSNR | SSIM | Consistency | Trust | Source |
|---|---|---|---|---|---|---|---|
| 1 | ScoreSPM + gradient | 0.761 | 30.52 | 0.924 | 0.93 | ✓ Certified | Wei et al., 2025 |
| 2 | E2E-BTR + gradient | 0.759 | 30.14 | 0.919 | 0.95 | ✓ Certified | Kossler et al., Sci. Rep. 2022 |
| 3 | U-Net-SPM + gradient | 0.753 | 30.65 | 0.926 | 0.88 | ✓ Certified | SPM U-Net variant |
| 4 | DiffusionSPM + gradient | 0.749 | 30.13 | 0.919 | 0.9 | ✓ Certified | Zhang et al., 2024 |
| 5 | SPM-Former + gradient | 0.729 | 29.29 | 0.905 | 0.87 | ✓ Certified | Chen et al., NanoLett 2024 |
| 6 | DeepSPM + gradient | 0.714 | 28.3 | 0.887 | 0.88 | ✓ Certified | Alldritt et al., Commun. Phys. 2020 |
| 7 | TV-Deconvolution + gradient | 0.651 | 24.67 | 0.792 | 0.94 | ✓ Certified | TV regularization for SPM |
| 8 | Reg-Deconv + gradient | 0.625 | 23.83 | 0.762 | 0.91 | ✓ Certified | Dongmo et al., 2000 |
| 9 | BTR + gradient | 0.585 | 22.19 | 0.698 | 0.92 | ✓ Certified | Villarrubia, JRNIST 1997 |
| 10 | MLE Reconstruction + gradient | 0.550 | 21.35 | 0.662 | 0.86 | ✓ Certified | Classical statistical method |
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%