Research Article
X-Ray Breast Images Denoising Method Based on the Convolutional Autoencoder
Table 2
The PSNR (dB) and SSIM of different methods at the MIAS dataset (k = 0.1, σ = 60).
| Different methods | SNR = 0.99 | SNR = 0.97 | SNR = 0.95 | SNR = 0.93 | PSNR | SSIM | PSNR | SSIM | PSNR | SSIM | PSNR | SSIM |
| Noise image | 21.97 | 0.21 | 19.22 | 0.14 | 17.55 | 0.10 | 16.32 | 0.07 | Literature [14] | 37.15 | 0.90 | 36.15 | 0.88 | 34.72 | 0.85 | 32.41 | 0.82 | Literature [15] | 36.82 | 0.89 | 35.64 | 0.87 | 33.48 | 0.84 | 31.52 | 0.81 | Literature [16] | 37.24 | 0.91 | 36.23 | 0.88 | 36.04 | 0.88 | 34.91 | 0.85 | Literature [17] | 37.29 | 0.91 | 36.59 | 0.89 | 36.21 | 0.88 | 35.73 | 0.87 | Proposed | 37.48 | 0.91 | 37.07 | 0.90 | 36.47 | 0.89 | 35.93 | 0.88 |
|
|