Table 3 Quantitative analysis results of the five models on the LSRV road dataset.

From: A dynamic attention mechanism for road extraction from high-resolution remote sensing imagery using feature fusion

Scheme

Network

Precision (%)

Rcall (%)

IoU (%)

F1-Score (%)

(a)

UNet

85.40

73.07

73.34

78.58

(b)

KanUNet

86.58

75.56

77.22

80.70

(c)

DenseUNet

86.84

76.63

78.33

81.34

(d)

MDCGANet

87.32

77.14

75.42

81.96

(e)

SCSM

87.96

77.51

77.19

82.42

(f)

RISENet

88.18

77.85

76.25

82.70

  1. Significant values are in [bold].
Back to article page