What is Convolutional Neural Network?

 Introduction

CNN is a convolutional neural network that uses neuron-based structure to analyze, process, and compute images for machine learning and data science applications.

Here below is the basic CNN network with basic components.

 

 

 

The basic components of the convolutional neural network:

1. Input Image data: We need raw data to process for neural network analysis.

 

2. Convolution Layer: This layer is used to correlate the images. In convolution, we flip the kernel and in correlation is do not flip the kernel. In a 3D image, the convolution layer converts it into a 2D image with the help of a filter with the same channels as in the 3D image. 

 

3. Pooling Layer: It further decreases the size of the image after the convolution layer. There are two types of options in pooling i.e. max pooling and average pooling.

 

4. Fully connected layer: It is a form of vector representation on the image after the pooling operation.

 

To further read more information about CNN with a python Example.

https://pub.towardsai.net/understand-cnn-basics-with-a-keras-example-in-python-c1fd6c449935?sk=6a4e45f9dd6b11e604311f2120b312c3