Table 1.

Confusion matrices: each column shows the reference, while numbers running diagonally show the percentage of correct classification for every class considered in the experimental study

Prediction		Reference
		1	2	3	4	> 4
VGG architecture
1		0.930	0	0	0	0
2		0.023	0.943	0.036	0.019	0.027
3		0.022	0.029	0.932	0.022	0.027
4		0.025	0.027	0.032	0.957	0.028
> 4		0	0	0	0.001	0.917
AlexNet architecture
1		0.910	0	0	0	0
2		0.030	0.944	0.039	0.030	0.026
3		0.029	0.029	0.920	0.032	0.026
4		0.031	0.027	0.041	0.937	0.031
> 4		0	0	0	0.001	0.917