Skip to main content
. 2022 Nov 3;10:958870. doi: 10.3389/fpubh.2022.958870

Table 1.

The full parameters of the adopted CNN structure.

No Layer Operation Type Kernel size Kernel amount Output
1 - - Input - - 100 × 100 × 3
2 L1 Convolution Conv2D 7 × 7 16 94 × 94 × 16
3 - Activation functions ReLu - - -
4 L2 Convolution Conv2D 5 × 5 32 90 × 90 × 32
5 - Activation functions ReLu -
6 L3 Fully connected layer FC - - 256
7 - Activation functions ReLu - - -
8 - Dropout layer Dropout - - -
9 L4 Fully connected layer FC - - 128
10 - Activation functions ReLu - - -
11 L5 Fully connected layer FC - - 6
13 L6 Activation functions Sigmoid - - 6