Table 12 Statistics of various indicators of recognition accuracy of RFCNN network model under different number of synthetic samples.

From: Analysis and construction of the coal and rock cutting state identification system in coal mine intelligent mining

Experimental grouping

Number of synthetic samples

Average value of recognition accuracy/%

Standard deviation of recognition accuracy

Variance of recognition accuracy

Average deviation of recognition accuracy

1

0

90.641

1.226 × 10−3

1.357 × 10−6

6.736 × 10−2

2

1000

93.262

3.258 × 10−4

1.205 × 10−7

3.284 × 10−3

3

2000

94.447

2.042 × 10−4

5.329 × 10−8

2.538 × 10−4

4

3000

96.193

1.577 × 10−4

1.969 × 10−8

2.303 × 10−4

5

4000

97.506

1.201 × 10−4

1.374 × 10−8

1.627 × 10−4

6

5000

98.344

1.714 × 10−5

2.883 × 10−10

1.115 × 10−5

7

6000

98.307

1.722 × 10−5

2.917 × 10−10

1.132 × 10−5

8

7000

98.295

1.718 × 10−5

2.904 × 10−10

1.127 × 10−5