Table 3 Comparison of detection accuracy between FACSNet and SRNet (Spatial domain) %
From: FACSNet: Forensics aided content selection network for heterogeneous image steganalysis
Dataset | Steganography algorithm | Detection model | 0.1 bpp | 0.2 bpp | 0.3 bpp | 0.4 bpp |
|---|---|---|---|---|---|---|
BOSSbase (All Original) | HILL | SRNet | 65.53 | 73.41 | 79.72 | 84.25 |
FACSNet | 65.53 | 73.41 | 79.72 | 84.25 | ||
HUGO | SRNet | 71.44 | 80.52 | 86.33 | 90.64 | |
FACSNet | 71.48 | 80.52 | 86.33 | 90.64 | ||
Heterogeneous-BOSSbase (Original + AI) | HILL | SRNet | 60.04 | 69.80 | 75.11 | 80.63 |
FACSNet | 65.48 | 73.41 | 79.68 | 84.21 | ||
HUGO | SRNet | 66.96 | 75.56 | 82.07 | 85.76 | |
FACSNet | 71.40 | 80.46 | 86.29 | 90.60 | ||
BOWS (All Original) | HILL | SRNet | 62.87 | 71.82 | 78.44 | 82.23 |
FACSNet | 62.87 | 71.82 | 78.44 | 82.25 | ||
HUGO | SRNet | 70.53 | 79.66 | 85.01 | 89.88 | |
FACSNet | 70.54 | 79.66 | 85.01 | 89.88 | ||
Heterogeneous-BOWS (Original + AI) | HILL | SRNet | 57.45 | 66.47 | 73.93 | 78.06 |
FACSNet | 62.77 | 71.73 | 78.40 | 82.22 | ||
HUGO | SRNet | 64.60 | 74.85 | 80.55 | 83.52 | |
FACSNet | 69.87 | 79.35 | 84.73 | 89.22 |
