Solar EUV/X-ray Imaging

Solar EUV/X-ray Imaging

Join the Competition Contribute to Benchmark

Standard reconstruction benchmark — forward model perfectly known, no calibration needed. Score = 0.5 × clip((PSNR−15)/30, 0, 1) + 0.5 × SSIM

# Method Score PSNR (dB) SSIM Source
🥇 DeepEM 0.707 29.95 0.916 ✓ Certified Su et al., ApJ 2022
🥈 SolarFormer 0.706 29.88 0.915 ✓ Certified SDO-based restoration, 2024
🥉 Pixon 0.652 27.83 0.877 ✓ Certified Pina & Puetter, PASP 1993
4 Richardson-Lucy 0.568 24.99 0.802 ✓ Certified Richardson 1972 / Lucy 1974

Dataset: PWM Benchmark (4 algorithms)

Blind Reconstruction Challenge — forward model has unknown mismatch, must calibrate from data. Score = 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

# Method Overall Score Public
PSNR / SSIM
 Dev
PSNR / SSIM
 Hidden
PSNR / SSIM
 Trust Source
🥇 SolarFormer + gradient 0.646
0.728
28.42 dB / 0.889
0.649
25.45 dB / 0.816
0.560
22.39 dB / 0.706
✓ Certified SDO-based restoration, 2024
🥈 Pixon + gradient 0.605
0.666
25.95 dB / 0.831
0.590
22.67 dB / 0.718
0.558
22.16 dB / 0.697
✓ Certified Pina & Puetter, PASP 1993
🥉 Richardson-Lucy + gradient 0.569
0.583
22.18 dB / 0.698
0.570
22.43 dB / 0.708
0.555
21.9 dB / 0.686
✓ Certified Richardson 1972 / Lucy 1974
4 DeepEM + gradient 0.559
0.700
27.46 dB / 0.869
0.541
21.17 dB / 0.653
0.435
17.98 dB / 0.499
✓ Certified Su et al., ApJ 2022

Complete score requires all 3 tiers (Public + Dev + Hidden).

Join the competition →
Scoring: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖) PSNR 40% · SSIM 40% · Consistency 20%
Public 3 scenes

Full-access development tier with all data visible.

What you get & how to use

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.

Public Leaderboard
# Method Score PSNR SSIM
1 SolarFormer + gradient 0.728 28.42 0.889
2 DeepEM + gradient 0.700 27.46 0.869
3 Pixon + gradient 0.666 25.95 0.831
4 Richardson-Lucy + gradient 0.583 22.18 0.698
Spec Ranges (4 parameters)
Parameter Min Max Unit
psf_degradation_(mirror_aging) -4.0 8.0 -
stray_light -1.0 2.0 -
flat_field_error -0.6 1.2 -
pointing_jitter -0.2 0.4 arcsec
View Details →
Dev 3 scenes

Blind evaluation tier — no ground truth available.

What you get & how to use

What you get: Measurements (y), ideal forward operator (H), and spec ranges only.

How to use: Apply your pipeline from the Public tier. Use consistency as self-check.

What to submit: Reconstructed signals and corrected spec. Scored server-side.

Dev Leaderboard
# Method Score PSNR SSIM
1 SolarFormer + gradient 0.649 25.45 0.816
2 Pixon + gradient 0.590 22.67 0.718
3 Richardson-Lucy + gradient 0.570 22.43 0.708
4 DeepEM + gradient 0.541 21.17 0.653
Spec Ranges (4 parameters)
Parameter Min Max Unit
psf_degradation_(mirror_aging) -4.8 7.2 -
stray_light -1.2 1.8 -
flat_field_error -0.72 1.08 -
pointing_jitter -0.24 0.36 arcsec
Submit Reconstruction View Details →
Hidden 3 scenes

Fully blind server-side evaluation — no data download.

What you get & how to use

What you get: No data downloadable. Algorithm runs server-side on hidden measurements.

How to use: Package algorithm as Docker container / Python script. Submit via link.

What to submit: Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

Hidden Leaderboard
# Method Score PSNR SSIM
1 SolarFormer + gradient 0.560 22.39 0.706
2 Pixon + gradient 0.558 22.16 0.697
3 Richardson-Lucy + gradient 0.555 21.9 0.686
4 DeepEM + gradient 0.435 17.98 0.499
Spec Ranges (4 parameters)
Parameter Min Max Unit
psf_degradation_(mirror_aging) -2.8 9.2 -
stray_light -0.7 2.3 -
flat_field_error -0.42 1.38 -
pointing_jitter -0.14 0.46 arcsec
Submit Algorithm View Details →

Blind Reconstruction Challenge

Challenge

Given measurements with unknown mismatch and spec ranges (not exact params), reconstruct the original signal. A method must be evaluated on all three tiers for a complete score. Scored on a composite metric: 0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖).

Input

Measurements y, ideal forward model H, spec ranges

Output

Reconstructed signal x̂

Spec DAG — Forward Model Pipeline

M → P → D

M Modulation
P Propagation
D Detector

Mismatch Parameters

Symbol Parameter Description Nominal Perturbed
p_d psf_degradation_(mirror_aging) PSF degradation (mirror aging) (-) 0.0 4.0
s_l stray_light Stray light (-) 0.0 1.0
f_e flat_field_error Flat-field error (-) 0.0 0.6
p_j pointing_jitter Pointing jitter (arcsec) 0.0 0.2

Credits System

40%
Platform Profit Pool
Revenue allocated to benchmark rewards
30%
Winner Share
Top algorithm receives from pool
$100
Min Withdrawal
Minimum payout threshold
Spec Primitives Reference (11 primitives)
P Propagation

Free-space or medium propagation kernel (Fresnel, Rayleigh-Sommerfeld).

M Mask / Modulation

Spatial or spatio-temporal amplitude modulation (coded aperture, SLM pattern).

Π Projection

Geometric projection operator (Radon transform, fan-beam, cone-beam).

F Fourier Sampling

Sampling in the Fourier / k-space domain (MRI, ptychography).

C Convolution

Shift-invariant convolution with a point-spread function (PSF).

Σ Summation / Integration

Summation along a physical dimension (spectral, temporal, angular).

D Detector

Sensor readout with gain g and noise model η (Gaussian, Poisson, mixed).

S Structured Illumination

Patterned illumination (block, Hadamard, random) applied to the scene.

W Wavelength Dispersion

Spectral dispersion element (prism, grating) with shift α and aperture a.

R Rotation / Motion

Sample or gantry rotation (CT, electron tomography).

Λ Wavelength Selection

Spectral filter or monochromator selecting a wavelength band.