Academia.eduAcademia.edu

Figure 2 – uploaded by Samet Memiş

See full PDFdownloadDownload figure
Fourthly, we give PSNR and SSIM results of DBA, MDBUTMF, BPDF, NAFSMF, DAMF, AWMF and ACmF for the image Cameraman with a noise density of 30%, in Figure 2. Fifthly, we then give PSNR and SSIM results of ACmF for the image Lena ranging in noise densities from 10% to 90%, in Figure 3.

Figure 2 Fourthly, we give PSNR and SSIM results of DBA, MDBUTMF, BPDF, NAFSMF, DAMF, AWMF and ACmF for the image Cameraman with a noise density of 30%, in Figure 2. Fifthly, we then give PSNR and SSIM results of ACmF for the image Lena ranging in noise densities from 10% to 90%, in Figure 3.

Related Figures (13)

Step 4. Convert A from double form to uint8 form
Step 4. Convert A from double form to uint8 form
Table 1. Mean PSNR results for the 15 traditional images with different SPN ratios Secondly, we give mean PSNR and MSSIM results of the methods mentioned above for 15 traditional test images with 512512 pixels (Baboon, Boat, Bridge, Cameraman, Elaine, Flintstones, Hill, House, Lake, Lena, Living Room, Parrot, Peppers, Pirate, and Plane) and 40 test images with 600x600 pixels in the TESTIMAGES Database [17] ranging in noise densities from 10% to 90%, in Table 1, 2, 3, and 4, respectively.
Table 1. Mean PSNR results for the 15 traditional images with different SPN ratios Secondly, we give mean PSNR and MSSIM results of the methods mentioned above for 15 traditional test images with 512512 pixels (Baboon, Boat, Bridge, Cameraman, Elaine, Flintstones, Hill, House, Lake, Lena, Living Room, Parrot, Peppers, Pirate, and Plane) and 40 test images with 600x600 pixels in the TESTIMAGES Database [17] ranging in noise densities from 10% to 90%, in Table 1, 2, 3, and 4, respectively.
Table 2. MSSIM results for the 15 traditional images with different SPN ratios Table 3. Mean PSNR results for the 40 images for TESTIMAGES Gallery with different SPN ratios
Table 2. MSSIM results for the 15 traditional images with different SPN ratios Table 3. Mean PSNR results for the 40 images for TESTIMAGES Gallery with different SPN ratios
Table 4. MSSIM results for the 40 images for TESTIMAGES Gallery with different SPN ratios
Table 4. MSSIM results for the 40 images for TESTIMAGES Gallery with different SPN ratios
Table 5. PSNR results of the methods for some traditional images with different SPN ratios
Table 5. PSNR results of the methods for some traditional images with different SPN ratios
able 8. Mean running time for the 40 images with different SPN ratios (in Seconds) Finally, we give the running time data of the algorithms in Table 7 and 8 for 15 traditional images and TESTIMAGES database with different SPN ratios, respectively. Here, we use MATLAB R2019a and a workstation with I(R) Xeon(R) CPU E5-1620 v4 @ 3.5 GHz and 64 GB RAM for these comparisons. The results show that ACmF outperforms the methods mentioned above except for DAME in terms of running time. Moreover, ACmF is much more successful NAFSMF and AWMF, which are known as successful in high noise densities.
able 8. Mean running time for the 40 images with different SPN ratios (in Seconds) Finally, we give the running time data of the algorithms in Table 7 and 8 for 15 traditional images and TESTIMAGES database with different SPN ratios, respectively. Here, we use MATLAB R2019a and a workstation with I(R) Xeon(R) CPU E5-1620 v4 @ 3.5 GHz and 64 GB RAM for these comparisons. The results show that ACmF outperforms the methods mentioned above except for DAME in terms of running time. Moreover, ACmF is much more successful NAFSMF and AWMF, which are known as successful in high noise densities.
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved