Skip to main content

Advertisement

Table 1 PSNR values of the reconstructed images using different ratios

From: Collaborating filtering using unsupervised learning for image reconstruction from missing data

Image Method 15% missing data 25% missing data 35% missing data
Lena FPC 46.2950 45.8888 45.6461
  ALM 48.2790 48.2248 45.5295
  IALM 48.2673 48.2278 46.0819
  PPG 45.0964 41.9579 40.2865
  PG 45.0966 41.9582 40.2863
  SVT with power method 48.3221 48.2745 48.2295
  Proposed approach 48.9848 48.7817 48.5952
Cameraman FPC 28.2855 32.3433 31.7894
  ALM 36.8075 33.9694 29.5169
  IALM 36.7101 33.9849 29.8495
  PPG 31.3222 27.9425 24.3797
  PG 31.3199 27.9458 24.3757
  SVT with power method 40.9528 40.5182 40.0158
  Proposed approach 43.3525 42.4823 40.4596
Flinstones FPC 28.9102 37.4826 37.2896
  ALM 42.9923 37.3583 34.3512
  IALM 42.8769 37.4573 34.6774
  PPG 31.5781 29.5989 24.3693
  PG 31.5936 29.5887 24.3708
  SVT with power method 43.6614 43.5614 43.2075
  Proposed approach 47.8234 47.3814 44.3920
House FPC 28.6407 27.7999 27.4826
  ALM 37.0989 36.4579 36.3583
  IALM 36.8952 36.5262 36.4573
  PPG 31.5992 29.8729 29.5998
  PG 31.6005 29.8769 29.5701
  SVT with power method 38.1457 38.0827 37.6494
  Proposed approach 39.0707 38.1743 37.8234
Man FPC 39.2493 38.5864 38.1565
  ALM 41.3640 41.2133 40.4455
  IALM 41.2952 41.2191 40.8616
  PPG 39.4997 39.1895 39.3266
  PG 39.5008 39.1889 39.3264
  SVT with power method 42.9572 42.4158 42.0357
  Proposed approach 44.3212 43.9939 43.6724
  1. For each test setting, six results are provided: FPC, ALM, IAM, PPG, PG, SVT with power method, and our proposed model. The best value of the PSNR is the number written in italics on each row