Classification of Very High Resolution Aerial Photos Using Spectral-Spatial Convolutional Neural Networks

Table 1

The summary of the CNN model layers.

Layer (type)	Output shape	Number of parameters
Input	(None, 3, 7, 7)	0
2D convolution	(None, 1, 5, 32)	2048
Max pooling	(None, 1, 2, 16)	0
Batch normalization	(None, 1, 2, 16)	64
Dropout	(None, 1, 2, 16)	0
Flatten	(None, 32)	0
Dense	(None, 32)	1056
Batch normalization	(None, 32)	128
Dropout	(None, 32)	0
Dense (Softmax)	(None, 7)	231