Lip Print Recognition Algorithm Based on Convolutional Network

Table 1

Information of all network layers.

Network layer	Name	Dimensions	Channel number	Step length	Activation function
C1	Convolution layer		8	1	ReLU
M1	Pool layer		8	2	——
C2	Convolution layer		32	1	ReLU
M2	Pool layer		32	2	——
C3	Convolution layer		64	1	ReLU
M3	Pool layer		64	2	——
C4	Convolution layer		128	1	ReLU
C5	Convolution layer		128	1	ReLU
M4	Pool layer		128	2	——
C6	Convolution layer		64	1	ReLU
M5	Pool layer		64	2	——
FC1	Fully connected layer		3136 nodes		ReLU
FC2	Fully connected layer		2048 nodes		ReLU
FC3	Fully connected layer		2048 nodes		ReLU
Softmax	Output layer		40 nodes		——