Skip to main content

Table 3 SSIM for additive Gaussian noise

From: A collaborative adaptive Wiener filter for image restoration using a spatial-domain multi-patch correlation model

Image Method SSIM
   σ η =10 σ η =20 σ η =30 σ η =40
Aerial Corrupted 0.9719 0.9046 0.8224 0.7388
491×434 NLM 0.9753 0.9058 0.8412 0.7872
  GLIDE-NLM 0.9747 0.9058 0.8427 0.7916
  PLOW 0.9734 0.9252 0.8760 0.8282
  BM3D 0.9783 0.9340 0.8835 0.8287
  AWF 0.9762 0.9304 0.8804 0.8322
  CAWF 0.9787 0.9351 0.8859 0.8365
Bridge Corrupted 0.9542 0.8512 0.7365 0.6305
5122 NLM 0.9611 0.8651 0.7925 0.7379
  GLIDE-NLM 0.9621 0.8757 0.8010 0.7363
  PLOW 0.9587 0.9051 0.8485 0.7943
  BM3D 0.9692 0.9126 0.8539 0.7963
  AWF 0.9652 0.9061 0.8477 0.7927
  CAWF 0.9676 0.9109 0.8523 0.7985
River Corrupted 0.9512 0.8429 0.7252 0.6196
(Kodak 11) NLM 0.9525 0.8343 0.7588 0.7042
768×512 GLIDE-NLM 0.9548 0.8461 0.7617 0.7007
  PLOW 0.9505 0.8732 0.7921 0.7299
  BM3D 0.9597 0.8792 0.8066 0.7437
  AWF 0.9574 0.8823 0.8142 0.7512
  CAWF 0.9609 0.8873 0.8207 0.7634
Bones Corrupted 0.9259 0.7735 0.6224 0.4979
512×768 NLM 0.9329 0.7946 0.7269 0.6821
  GLIDE-NLM 0.9379 0.8163 0.7351 0.6725
  PLOW 0.9240 0.8295 0.7587 0.7047
  BM3D 0.9417 0.8404 0.7641 0.7091
  AWF 0.9389 0.8459 0.7669 0.6970
  CAWF 0.9439 0.8551 0.7839 0.7277
Building Corrupted 0.9153 0.7660 0.6349 0.5318
768×512 NLM 0.9395 0.8370 0.7808 0.7386
  GLIDE-NLM 0.9426 0.8416 0.7835 0.7385
  PLOW 0.9379 0.8577 0.7823 0.7262
  BM3D 0.9482 0.8748 0.8141 0.7637
  AWF 0.9398 0.8583 0.7797 0.7072
  CAWF 0.9502 0.8788 0.8190 0.7665
Gazebo Corrupted 0.9130 0.7678 0.6428 0.5435
768×512 NLM 0.9561 0.8653 0.8090 0.7649
  GLIDE-NLM 0.9578 0.8712 0.8117 0.7686
  PLOW 0.9525 0.8698 0.8046 0.7559
  BM3D 0.9620 0.8963 0.8397 0.7940
  AWF 0.9452 0.8740 0.7950 0.7230
  CAWF 0.9611 0.8988 0.8387 0.7837