Skip to main content

Table 7 Performance indices obtained by training on the LIVE or TID2013 database and testing on the TID2013 or LIVE database

From: No-reference color image quality assessment: from entropy to perceptual quality

Method Trained on LIVE, tested on TID2013 Trained on TID2013, tested on LIVE
  SROCC PLCC RMSE SROCC PLCC RMSE
PSNR 0.9244 0.9140 5.6710 0.8850 0.8825 12.7229
SSIM 0.8662 0.8723 6.8356 0.9255 0.9241 10.3565
VIF 0.9181 0.9401 4.7645 0.9287 0.9318 9.8436
BIQI 0.8388 0.8447 7.4817 0.5699 0.6315 22.4137
DIIVINE 0.7958 0.7868 8.6283 0.8012 0.8803 15.0072
BLIINDS-II 0.8503 0.8413 7.5559 0.9267 0.9476 10.3571
BRISQUE 0.8817 0.8860 6.4821 0.8969 0.8962 14.1317
NIQE 0.6798 0.6685 10.3965 0.9287 0.9447 10.7151
ILNIQE 0.7793 0.7956 8.4692 0.8537 0.8935 12.9788
SSEQ 0.8294 0.8769 6.7188 0.8827 0.9167 11.5529
ENIQA 0.8916 0.8953 6.2283 0.9026 0.9093 12.2189
  1. The top performances of the FR-IQA and NR-IQA indices are highlighted in italics. The second-best results of the NR-IQA indices are boldfaced