Skip to main content

Table 1 Performance comparison of Xiamen, China

From: Land cover classification combining Sentinel-1 and Landsat 8 imagery driven by Markov random field with amendment reliability factors

Method

Precision

Urban

water

Vegetation

Overall accuracy

Kappa

MRF-ARF

PA

97.87%

85.20%

87.47%

93.61%

0.8717

UA

92.46%

87.96%

96.40%

MRF-RF

PA

58.48%

88.81%

62.45%

61.00%

0.4209

UA

97.99%

8.48%

84.01%

ATWT-EMD

PA

66.48%

88.65%

60.02%

64.90%

0.3549

UA

74.38%

29.93%

60.98%

SAR only

PA

95.55%

37.16%

38.87%

72.75%

0.3938

UA

70.52%

99.76%

81.07%

Optical only

PA

59.25%

93.26%

77.50%

67.15%

0.4444

UA

96.97%

25.77%

56.05%