A convolution layer is mainly characterized by its number
of convolution filters and the size of these filters. Each filter
moves within the image (the filter has to be within the image
completely) to generate a convolved image. The output of
a convolution layer hij , which is also called “feature map,”
is a pixelwise transformation of the convolved images by an
activation function.With multiple filters, multiple feature maps
are generated and stacked to form a multilayer featuremap. The
convolution process for each filter is described