Research Article
Zero-Watermarking Algorithm for Medical Image Based on VGG19 Deep Convolution Neural Network
Table 7
Experimental results under rotation distortion attack.
| ā | Distortion quantity (%) | 10 | 30 | 50 | 70 | 90 |
| Medical image A | PSNR (dB) | 13.8738 | 13.2751 | 12.8138 | 12.5202 | 12.2938 | NC | 1.0 | 1.0 | 1.0 | 1.0 | 0.94054 |
| Medical image B | PSNR (dB) | 21.5541 | 16.336 | 14.1869 | 12.7102 | 11.5936 | NC | 0.93883 | 0.81049 | 0.87509 | 0.62698 | 0.62927 |
| Medical image C | PSNR (dB) | 23.2593 | 16.2611 | 14.4212 | 13.6166 | 13.1196 | NC | 0.93626 | 0.93626 | 0.93626 | 0.93626 | 0.5621 |
| Medical image D | PSNR (dB) | 23.6471 | 17.9937 | 16.1955 | 15.1909 | 14.7219 | NC | 0.94054 | 0.94054 | 0.90682 | 0.90682 | 0.90682 |
| Medical image E | PSNR (dB) | 32.9927 | 27.8346 | 25.4521 | 23.7641 | 22.4491 | NC | 0.9394 | 0.9394 | 0.87823 | 0.87823 | 0.8182 |
|
|