Table 1 Comparison experiment of feature extraction methods on dataset 1

From: Automatic labeling framework for paint loss disease of ancient murals based on hyperspectral image classification and segmentation

Combination

PCA

MNF

DCT-ICA

PA(%)

1

\(\surd \)

\(\times \)

\(\times \)

33.85

2

\(\times \)

\(\surd \)

\(\times \)

68.47

3

\(\times \)

\(\times \)

\(\surd \)

55.34

4

\(\surd \)

\(\surd \)

\(\times \)

83.65

5

\(\surd \)

\(\times \)

\(\surd \)

66.79

6

\(\times \)

\(\surd \)

\(\surd \)

86.34

7

\(\surd \)

\(\surd \)

\(\surd \)

93.16

  1. Bold represents the best result
Back to article page