Blotch and scratch removal in archived film using a semi-transparent corruption model and a ground-truth generation technique

EURASIP Journal on Image and Video Processing

Table 1 Reconstruction quality against ground truth for the examined blotch sequences

System	LabB1	DanceB1	DanceB2	ArtB1
Furnace
Average	0.76 ± 0.08	0.83 ± 0.04	0.80 ± 0.06	0.77 ± 0.04
Minimum	0.44	0.64	0.64	0.69
Maximum	0.93	0.98	0.89	0.98
JONDI
Average	0.98 ± 0.01	0.94 ± 0.01	0.90 ± 0.02	0.98 ± 0.00
Minimum	0.94	0.89	0.84	0.96
Maximum	0.99	0.99	0.94	0.99
BTBR-AOFF
Average	0.84 ± 0.07	0.90 ± 0.02	0.90 ± 0.04	0.88 ± 0.04
Minimum	0.70	0.84	0.78	0.82
Maximum	0.94	0.94	0.95	0.95
BBM
Average	0.89 ± 0.03	0.94 ± 0.01	0.94 ± 0.02	0.94 ± 0.01
Minimum	0.82	0.91	0.86	0.93
Maximum	0.95	0.96	0.98	0.97
BTBR-F
Average	0.97 ± 0.01	0.96 ± 0.01	0.95 ± 0.02	0.98 ± 0.01
Minimum	0.93	0.93	0.89	0.97
Maximum	0.99	0.98	0.98	0.99

SSIM [50] is used here where an SSIM of 1 means that the sequence is identical to the ground truth reconstruction. The average and the minimum SSIM show that BTBR-F is the best performing system.