The architecture of a CNN is shown in the following diagram:
As you will notice, first we feed the input image to the convolutional layer, where we apply the convolution operation to extract important features from the image and create the feature maps. We then pass the feature maps to the pooling layer, where the dimensions of the feature maps will be reduced. As shown in the previous diagram, we can have multiple convolutional and pooling layers, and we should also note that the pooling layer does not necessarily have to be there after every convolutional layer; there can be many convolutional layers followed by a pooling layer.
So, after the convolutional and pooling layers, we flatten the resultant feature maps and feed it to a fully connected layer, which is basically a feedforward neural network that classifies the given input image based on the...