Introduction

In the previous chapters, we discussed Dense Neural Networks (DNNs) in which each neuron of a layer is connected to each neuron of the adjacent layer. In this chapter, we will focus on a special type of neural network that performs well for image classification: CNNs.

A CNN is a combination of two components: a feature extractor module followed by a trainable classifier. The first component includes a stack of convolution, activation, and pooling layers. A DNN does the classification. Each neuron in a layer is connected to those in the next layer.

In mathematics, a convolution is a function that is applied over the output of another function. In our case, we will consider using a matrix multiplication (filter) across an image. For our purposes, we find an image to be a matrix of numbers. These numbers may represent pixels or even image attributes. The convolution operation we will apply to these matrices involves moving a filter of fixed width across...