Table 6 Structural parameters of generator and discriminator models.

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

Generator

Enter channel dimension

Output channel dimension

Input dimension

Output dimension

Convolution kernel dimension

Step

deconv1

512

384

4*4

8*8

3*3

2

conv-upsample1

384

128

8*8

22*22

3*3

3

deconv2

384

128

8*8

22*22

3*3

3

concact1

Feature connection

256

Feature connection

22*22

\

\

deconv3

256

48

22*22

43*43

5*5

2

conv-upsample2

256

48

22*22

43*43

5*5

2

concact2

Feature connection

96

Feature connection

43*43

\

\

deconv4

96

16

43*43

128*128

11*11

3

conv-upsample3

96

16

43*43

128*128

11*11

3

concact3

Feature connection

32

Feature connection

128*128

\

\

fc

32

3

128*128

128*128

1*1

1

Discriminator

Enter channel dimension

Output channel dimension

Input dimension

Output dimension

Convolution kernel dimension

Step

conv1

3

64

128*128

43*43

7*7

3

conv2

64

128

43*43

22*22

5*5

2

conv3

128

256

22*22

8*8

3*3

3

conv4

256

512

8*8

4*4

3*3

2

maxpooling

256

256

8*8

4*4

2*2

2

maxpooling

256

256

8*8

4*4

2*2

2

Back to article page