Deep Convolutional Asymmetric Autoencoder-Based Spatial-Spectral Clustering Network for Hyperspectral Image

Table 1

Network parameter settings of CAE.

Layers	Kernel	Output	Channel
Input layer	/	55	204
Conv layer	(3, 3) 256	33	256
Conv layer	(3, 3) 512	11	512
Dense layer	/	512	/
Dense layer	/	256	/
Dense layer	/	128	/
Dense layer	/	256	/
Dense layer	/	512	/
Reshape layer	/	11	512
Transpose layer	(3, 3)^T 512	33	512
Transpose layer	(3, 3)^T 256	55	256
Output layer	(3, 3) 204	55	204