# Input Layer and Convolution Layer

Understand the fundamentals of input layers and convolution layers.

In this lesson, we’ll focus on the fundamental components of CNNs, specifically the input and convolution layers. These layers play a crucial role in processing and analyzing image data. The input layer receives raw image data, while the convolution layer extracts important features using different kernels. By exploring the functionalities and parameters of these layers, we can understand how CNNs are effective at image-related tasks.

## Input layer

The input layer is the first layer of a CNN. Its primary role is to receive the input data, usually in the form of images, and transmit it through the network for further processing. The input layer is specifically designed to handle the spatial characteristics of the input data. It keeps the dimensions and arrangement of the input intact, ensuring the preservation of the spatial relationships between the pixels in an image.

## Role of the convolution layer in CNNs

The convolution layer is a fundamental building block of a CNN that performs extensive calculations. It uses a special operation called convolution, which involves multiplying two matrices and adding up the results. One matrix, called the kernel or filter, has some numbers that can be changed. The other matrix is a small part of the input image. When we multiply and add everything up, we obtain the convolution result.

This process of convolution (for 2D data) can be mathematically written as the following:

Get hands-on with 1200+ tech skills courses.