Public

Raman Imaging / Microscopy — 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
spectral_calibration_shift -0.4 – 0.8 0.2 cm^-1
fluorescence_background -2.0 – 4.0 1.0 relative
laser_power_fluctuation -1.0 – 2.0 0.5 -
cosmic_ray_artifact -0.2 – 0.4 0.1 -

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

12.37 dB

SSIM 0.2289

Scenario II (Mismatch)

11.10 dB

SSIM 0.0825

Scenario III (Oracle)

12.54 dB

SSIM 0.0863

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 11.90 0.1867 10.36 0.0587 10.94 0.0808
scene_01 13.20 0.1668 12.09 0.0565 12.66 0.0783
scene_02 11.29 0.2707 9.97 0.1188 12.91 0.0879
scene_03 13.08 0.2912 11.97 0.0959 13.66 0.0985

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 U-Net-Spectra + gradient 0.758 30.24 0.92 0.94 ✓ Certified Spectral U-Net variant
2 PINN-Spectra + gradient 0.758 30.57 0.925 0.91 ✓ Certified Physics-informed neural network
3 Cascade-UNet + gradient 0.756 30.87 0.929 0.88 ✓ Certified Physics-informed UNet, 2025
4 DiffusionSpectra + gradient 0.747 29.52 0.909 0.94 ✓ Certified Zhang et al., 2024
5 CDAE + gradient 0.730 29.19 0.904 0.88 ✓ Certified Zhang et al., Sensors 2024
6 ScoreSpectra + gradient 0.720 28.64 0.894 0.88 ✓ Certified Wei et al., 2025
7 SpectraFormer + gradient 0.710 27.75 0.875 0.91 ✓ Certified Spectroscopy transformer, 2024
8 PnP-DnCNN + gradient 0.659 25.44 0.816 0.89 ✓ Certified Zhang et al., 2017
9 SVD + gradient 0.633 24.58 0.789 0.86 ✓ Certified Singular Value Decomposition
10 Baseline Correction + gradient 0.580 22.44 0.708 0.86 ✓ Certified Polynomial fitting baseline
11 SG-ALS + gradient 0.565 21.61 0.673 0.9 ✓ Certified Savitzky-Golay + ALS baseline

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 Raman Imaging / Microscopy