TABLE 4.

The hyper parameters of the proposed BC-DA-RCNN model.

	Layer type	Size	Stride	Output shape
1	Input	32 × 32	–	None
2	Convolution layer	32 filters size of (3 × 3 or 5 × 5 or 7 × 7)	1	32 × 32 × 32
3	Residual block 1	32 filters size of (3 × 3 or 5 × 5 or 7 × 7)	1	32 × 32 × 32
4	Residual block 2	64 filters size of (3 × 3 or 5 × 5 or 7 × 7)	1	32 × 32 × 64
5	Residual block 3	128 filters size of (3 × 3 or 5 × 5 or 7 × 7)	1	32 × 32 × 128
6	Dense layer 1	1,024 units with dropout rate: 0.2	–	1024
7	Dense layer 2	512 units with dropout rate: 0.2	–	512
8	SoftMax	–	–	4 or 2