Public

Sonar — 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
sound_speed_profile -5.0 – 10.0 2.5 m/s
multipath -2.0 – 4.0 1.0
array_calibration -3.0 – 6.0 1.5 %

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

1.16 dB

SSIM 0.0122

Scenario II (Mismatch)

1.16 dB

SSIM 0.0118

Scenario III (Oracle)

4.17 dB

SSIM 0.5287

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 1.17 0.0105 1.17 0.0100 4.11 0.5182
scene_01 1.02 0.0148 1.01 0.0145 4.12 0.5434
scene_02 1.28 0.0131 1.28 0.0125 4.35 0.5346
scene_03 1.17 0.0105 1.17 0.0101 4.11 0.5184

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 AcousticFormer + gradient 0.780 31.9 0.942 0.92 ✓ Certified Acoustic imaging transformer, 2024
2 SonarNet + gradient 0.696 26.62 0.849 0.95 ✓ Certified Underwater imaging CNN, 2022
3 MVDR/Capon + gradient 0.563 21.82 0.682 0.86 ✓ Certified Capon, Proc. IEEE 1969
4 DAS + gradient 0.524 20.43 0.619 0.86 ✓ Certified Van Trees, Array Processing, 2002

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 Sonar