Hidden

Dark-Field Microscopy — Hidden Tier

(5 scenes)

Fully blind server-side evaluation — no data download.

No data downloadable. Algorithm runs server-side on hidden measurements.

Package algorithm as Docker container / Python script. Submit via link.

Containerized algorithm accepting y + H, outputting x_hat + corrected spec.

🔒

True spec hidden — blind evaluation, only ranges available.

Parameter	Spec Range	Unit
condenser_na_vs_objective_na_ratio	1.158 – 1.338	-
stray_light	-0.7 – 2.3	relative
scattering_angle_range	-0.15 – 0.15	-

#	Method	Score	PSNR	SSIM	Consistency	Trust	Source
1	DiffusionDF + gradient	0.743	31.01	0.931	0.8	✓ Certified	Luo et al., arXiv 2023 (DF)
2	SwinIR-DF + gradient	0.716	28.53	0.891	0.87	✓ Certified	Liang et al., ICCV 2021 (DF)
3	BM3D-DF + gradient	0.707	28.83	0.897	0.8	✓ Certified	Dabov et al., IEEE TIP 2007 (DF adapt.)
4	Restormer-DF + gradient	0.706	28.99	0.9	0.78	✓ Certified	Zamir et al., CVPR 2022 (DF)
5	Wiener-DF + gradient	0.600	24.02	0.769	0.76	✓ Certified	Wiener, 1949 (DF adapt.)
6	Noise2Void-DF + gradient	0.575	22.12	0.695	0.88	✓ Certified	Krull et al., CVPR 2019 (DF)
7	Richardson-Lucy + gradient	0.552	21.41	0.664	0.86	✓ Certified	Richardson, JOSA 1972; Lucy, AJ 1974
8	CARE-DF + gradient	0.540	21.8	0.681	0.75	✓ Certified	Weigert et al., Nat. Methods 2018 (DF)
9	TV-DF + gradient	0.408	16.47	0.424	0.87	✓ Certified	Rudin et al., Physica D 1992 (DF)

Scenes: 5

0.4 × PSNR_norm + 0.4 × SSIM + 0.2 × (1 − ‖y − Ĥx̂‖/‖y‖)

PSNR: 40% SSIM: 40% Consistency: 20%