Skip to main content

Advertisement

Table 3 Perceptual quality measure results

From: A supervised visual model for finding regions of interest in basal cell carcinoma images

  Itti's model Achanta's model Proposed model
  Train (dB) Test (dB) Train (dB) Test (dB) Train (dB) Test (dB)
False fore-ground -57.3 ± 0.2 -56.0 ± 2.0 -60.6 ± 0.2 -59.3 ± 1.9 -52.7 ± 0.3 -51.2 ± 3.2
Added back- ground -58.7 ± 0.1 -57.3 ± 1.4 -50.8 ± 0.4 -48.8 ± 5.5 -52.0 ± 0.1 -51.6 ± 1.2
Total -61.8 ± 0.1 -60.3 ± 1.4 -63.0 ± 0.1 -61.6 ± 1.1 -57.7 ± 0.1 -56.7 ± 1.6
  1. Average ± standard deviation of the perceptual quality in false foreground, added background and total quality for proposed and Itti's visual attention models obtained on the training and test sets.