Skip to main content

Table 5 Recognition accuracy (%) on the Cohn-Kanade database for different approaches

From: Gauss–Laguerre wavelet textural feature fusion with geometrical information for facial expression identification

Expression

Leave-One-Out

Cross validation

Expresser-based segmentation

Anger

88.25

87.03

82.35

Disgust

94.9

91.58

88.65

Fear

92.28

90.98

87.12

Happiness

97.82

96.92

91.05

Sad

87.87

84.58

81.11

Surprise

92.08

91.23

86.32

Average

92.2

90.37

86.1