Table 4 Performance comparisons with baselines on the YouTube-w/o-ALI dataset, unit: %.

From: Transferring enhanced material knowledge via image quality enhancement and feature distillation for pavement condition identification

Model

Acc.

Rec.

Spec.

CNN-3

90.96

88.21

88.65

CNN-2

90.16

88.98

89.21

CNN-1

89.96

85.29

85.78

SqueezeNet

93.59

92.36

92.48

VGG16

91.65

90.58

90.87

VGG19

91.79

91.44

91.88

ResNet50

92.17

92.30

92.51

ResNeXt50

93.36

94.06

97.08

EfficientNet-B3

94.58

94.76

97.23

FocalNet

95.16

95.06

97.85

DenseNet121

95.46

96.02

96.25

MBFN

97.71

97.82

97.34

IQEFD (Our)

98.68

98.90

98.65

  1. The best value of each metric is bold and italic.
Back to article page