Table 2.

Using the pre-trained convolutional neural network (CNN) models as a feature extractor

Structure	Convolutional neural network
Input images (107 × 107)	Conv layer 1	Conv layer 2	Conv layer 3
	64 × 11 × 11	256 × 5 × 5	256 × 3 × 3
	str. 4, pad 0, lrn, *2 pool	str. 1, pad 0, lrn, *2 pool	str. 1, pad 0