|
Figure 7
CNN representations of the first convolutional layers of AlexNet with two distinct classes (Strong and No crystal). The visual representations indicate that the network extracts uniquely different representations for the two classes. |
Open
access
