Table 10 Class-wise Score of KSC dataset using D-Path-AE.

From: Enhancing feature learning of hyperspectral imaging using shallow autoencoder by adding parallel paths encoding

Number

Name

SVM

KNN

DT

1

scrub

42.01

88.40

88.15

2

willow swamp

0.03

78.41

75.21

3

cabbage palm hammock

0.02

71.03

83.21

4

cabbage palm/oak hammock

0.12

36.33

41.21

5

slash pine

0.21

60.11

60.76

6

oak/broadleaf hammock

0.03

41.43

38.62

7

hardwood swamp

0.42

67.91

62.33

8

graminoid marsh

0.21

68.00

69.41

9

spartina marsh

11.21

82.10

84.21

10

cattail marsh

9.82

78.44

78.90

11

salt marsh

0.03

96.11

95.22

12

mud flats

5.03

84.60

80.00

13

water

71.51

98.21

98.00

Kappa

 

22.48

78.15

77.63

OA

 

35.44

80.45

79.88

AA

 

16.37

72.40

73.78