Deep Learning Technology for Weld Defects Classification Based on Transfer Learning and Activation Features

Table 1

The principal parameters of the CNN design.

1	Data	Image input	images with zero center normalization
2	Conv1	Convolution	96 convolutions with stride [4, 4] and padding [0, 0, 0, 0]
3	Relu1	ReLU	ReLU
4	Norm1	Cross channel normalization	Cross channel normalization with 5 channels per element
5	Pool1	Max pooling	max pooling with stride [2, 2] and padding [0, 0, 0, 0]
6	Conv2	Convolution	256 4 convolutions with stride [1, 1] and padding [2, 2, 2, 2]
7	Relu2	ReLU	ReLU
8	Norm2	Cross channel normalization	Cross channel normalization with 5 channels per element
9	Pool2	Max pooling	max pooling with stride [2, 2] and padding [0, 0, 0, 0]
10	Conv3	Convolution	348 convolutions with stride [1, 1] and padding [1, 1, 1, 1]
11	Relu3	ReLU	ReLU
12	Con4	Convolution	348 convolutions with stride [1, 1] and padding [1, 1, 1, 1]
13	Relu4	ReLU	ReLU
14	Con5	Convolution	256 convolutions with stride [1, 1] and padding [1, 1, 1, 1]
15	Relu5	ReLU	ReLU
16	Pool5	Max pooling	max pooling with stride [2, 2] and padding [0, 0, 0, 0]
17	FC6	Fully connected	4096 fully connected layers
18	Relu6	ReLU	ReLU
19	Drop6	Dropout	50% dropout
20	FC7	Fully connected	4096 fully connected layers
21	Relu7	ReLU	ReLU
22	Drop7	Dropout	50% dropout
23	FC8	Fully connected	2 fully connected layers
24	Prob	Softmax	Softmax
25	Output	Classification output	Cross entropy