Table 9.

Confusion matrix on the test set of the VGG19 network with weights initialized on ImageNet, and trained using the mean square error loss function and the Adam optimizer

	W	N	SN	SH	S	P	IF	BH	F	FA
W	100.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
N	0.0	97.3	2.4	0.0	0.0	0.0	0.0	0.3	0.0	0.0
SN	0.0	0.2	96.8	1.9	0.0	0.9	0.0	0.0	0.2	0.0
SH	1.5	0.0	3.3	88.0	0.0	1.6	0.0	5.2	0.4	0.0
S	7.0	0.0	0.0	0.0	92.3	0.0	0.0	0.0	0.0	0.7
P	0.0	0.0	0.2	0.1	0.0	96.7	0.4	2.5	0.0	0.0
IF	0.0	0.0	0.0	0.1	0.0	3.9	94.4	1.5	0.0	0.0
BH	0.0	0.0	0.0	3.6	0.6	1.1	0.0	91.7	2.5	0.7
F	0.0	0.0	0.0	0.4	0.0	1.4	1.6	0.0	96.6	0.0
FA	0.0	0.0	0.0	0.0	0.0	0.7	0.0	0.7	0.0	98.6

Results are obtained setting the size of the sliding window equal to 30 and using the weighted sum (WS) aggregation rule