Public

Ground-Penetrating Radar (GPR) — 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
soil_permittivity_error 7.8 – 11.4 9.6 -
antenna_height -0.01 – 0.02 0.005 m
time_zero_offset -0.1 – 0.2 0.05 ns
velocity_model_error 0.094 – 0.112 0.103 m/ns

InverseNet Baseline Scores

Method: CPU_baseline — Mismatch parameter: nominal

Scenario I (Ideal)

6.10 dB

SSIM 0.0513

Scenario II (Mismatch)

6.10 dB

SSIM 0.0506

Scenario III (Oracle)

9.04 dB

SSIM 0.5406

Per-scene breakdown (4 scenes)
Scene PSNR I SSIM I PSNR II SSIM II PSNR III SSIM III
scene_00 6.10 0.0509 6.10 0.0505 9.04 0.5407
scene_01 6.10 0.0513 6.10 0.0507 9.04 0.5407
scene_02 6.10 0.0517 6.10 0.0507 9.04 0.5406
scene_03 6.10 0.0515 6.10 0.0506 9.04 0.5406

Public Tier Leaderboard

# Method Score PSNR SSIM Consistency Trust Source
1 HyperDet + gradient 0.729 28.95 0.899 0.9 ✓ Certified GPR detection transformer, 2023
2 GPR-RCNN + gradient 0.695 27.01 0.858 0.91 ✓ Certified Pham & Lefevre, JECE 2020
3 RTM + gradient 0.608 23.35 0.745 0.88 ✓ Certified Baysal et al., Geophysics 1983
4 Kirchhoff Migration + gradient 0.494 19.06 0.553 0.91 ✓ Certified Stolt, Geophysics 1978

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 Ground-Penetrating Radar (GPR)