Table 1.

CNN configuration.

Name	Type	Activations	Learnables
Input	Image input	25 × 25 × 1	-
Conv_1	Convolution	23 × 23 × 8	Weights 3 × 3 × 1 × 8 and Bias 1 × 1 × 8
Relu1	Rectified linear unit	23 × 23 × 8	-
2 × 2-MP	Max pooling	11 × 11 × 8	-
Conv_2	Convolution	9 × 9 × 8	Weights 3 × 3 × 8 × 8 and Bias 1 × 1 × 8
Relu2	Rectified linear unit	9 × 9 × 8	-
FC	Fully connected	1 × 1 × 4	Weights 4 × 648 and Bias 4 × 1
SM	Softmax	1 × 1 × 4	-
Class	Classification output	-	-