Public

Talbot-Lau X-ray Grating Interferometry — Public Tier

(5 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
grating_alignment_(rotation) -0.1 – 0.2 0.05 deg
inter_grating_distance_error -0.2 – 0.4 0.1 -
phase_stepping_error -1.0 – 2.0 0.5 perstep
grating_defect_fraction -0.6 – 1.2 0.3 -

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

22.61 dB

SSIM 0.7075

Scenario II (Mismatch)

20.21 dB

SSIM 0.1585

Scenario III (Oracle)

23.95 dB

SSIM 0.1525

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 23.97 0.7060 21.29 0.1538 24.36 0.1416
scene_01 22.69 0.6936 20.16 0.1525 23.95 0.1560
scene_02 21.95 0.7536 20.50 0.1715 23.95 0.1510
scene_03 21.82 0.6767 18.89 0.1561 23.54 0.1613

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 GratingFormer + gradient 0.741 30.04 0.917 0.87 ✓ Certified Grating interferometry transformer, 2024
2 DPC-Net + gradient 0.716 27.95 0.88 0.92 ✓ Certified Differential phase contrast CNN, 2021
3 PCA Retrieval + gradient 0.560 21.34 0.661 0.91 ✓ Certified Zanette et al., PMB 2012
4 Phase Stepping + gradient 0.557 21.54 0.67 0.87 ✓ Certified Weitkamp et al., Opt. Express 2005

Visible Data Fields

y H_ideal spec_ranges x_true true_spec

Dataset

Format: HDF5
Scenes: 5

Scoring Formula

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%
Back to Talbot-Lau X-ray Grating Interferometry