Table 4 Classification accuracy (percentage) for varying parameter k with linear SVM and nonlinear SVM as classification kernel.
From: A Novel Image Data Hiding Scheme with Diamond Encoding
Parameter | Linear SVM | Nonlinear SVM | ||
|---|---|---|---|---|
k | Stego-image | Clean image | Stego-image | Clean image |
1 | 11.3 | 89.4 | 13.1 | 90.5 |
2 | 22.7 | 83.6 | 26.3 | 87.3 |
3 | 30.2 | 79.5 | 35.2 | 85.2 |
4 | 32.7 | 76.8 | 41.3 | 80.5 |