Public

Industrial CT — 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
beam_hardening_order -0.5 – 1.0 0.25
scatter_fraction -0.15 – 0.3 0.075
source_blur -3.0 – 6.0 1.5 pixels
detector_efficiency 0.7 – 1.15 0.925

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

7.34 dB

SSIM 0.0390

Scenario II (Mismatch)

7.02 dB

SSIM 0.0210

Scenario III (Oracle)

10.82 dB

SSIM 0.0936

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 7.34 0.0377 6.96 0.0206 10.91 0.0947
scene_01 7.43 0.0387 7.16 0.0206 10.70 0.0875
scene_02 7.18 0.0422 6.94 0.0219 10.69 0.1025
scene_03 7.40 0.0373 7.02 0.0210 10.97 0.0896

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 Learned Primal-Dual + gradient 0.803 33.86 0.96 0.9 ✓ Certified Adler & Oktem, IEEE TMI 2018
2 FBPConvNet + gradient 0.795 33.58 0.958 0.88 ✓ Certified Jin et al., IEEE TIP 2017
3 PnP-ADMM + gradient 0.748 30.22 0.92 0.89 ✓ Certified Venkatakrishnan et al., 2013
4 FDK + gradient 0.731 28.48 0.891 0.95 ✓ Certified Feldkamp et al., JOSA A 1984

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%
Back to Industrial CT