Public

Ocean Color Remote Sensing — 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
atmospheric_correction_error -3.0 – 6.0 1.5 -
sun_glint_contamination -4.0 – 8.0 2.0 -
vicarious_calibration_offset -0.6 – 1.2 0.3 -

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

9.15 dB

SSIM 0.0293

Scenario II (Mismatch)

9.23 dB

SSIM 0.0312

Scenario III (Oracle)

12.18 dB

SSIM 0.0671

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 9.16 0.0297 9.24 0.0305 12.19 0.0654
scene_01 9.15 0.0294 9.20 0.0320 12.16 0.0681
scene_02 9.12 0.0284 9.25 0.0314 12.20 0.0668
scene_03 9.15 0.0298 9.20 0.0311 12.16 0.0679

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 AquaFormer + gradient 0.751 30.61 0.926 0.87 ✓ Certified Ocean color retrieval transformer, 2024
2 OC-Net + gradient 0.709 27.8 0.877 0.9 ✓ Certified Pahlevan et al., RSE 2022
3 MUMM + gradient 0.612 23.37 0.745 0.9 ✓ Certified Ruddick et al., RSE 2000
4 Gordon AC + gradient 0.557 21.03 0.647 0.94 ✓ Certified Gordon & Wang, Appl. Opt. 1994

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 Ocean Color Remote Sensing