Scribd
Upload
104 views4 pages

Theory of CNN (Convolutional Neural Network)

The document discusses the theory of Convolutional Neural Networks (CNNs). It defines CNNs as neural networks that can process grid-like data such as images. CNNs employ mathematical operations like convolution and pooling layers to extract features from input data. The convolutional layer uses filters to convolve the input and extract features, while the pooling layer reduces the spatial size of representations to control overfitting. Together, the convolutional and pooling layers form the basic building block of CNN architectures.

Uploaded by

deepak mauu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
104 views4 pages

Theory of CNN (Convolutional Neural Network)

The document discusses the theory of Convolutional Neural Networks (CNNs). It defines CNNs as neural networks that can process grid-like data such as images. CNNs employ mathematical operations like convolution and pooling layers to extract features from input data. The convolutional layer uses filters to convolve the input and extract features, while the pooling layer reduces the spatial size of representations to control overfitting. Together, the convolutional and pooling layers form the basic building block of CNN architectures.

Uploaded by

deepak mauu
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Theory of CNN

(Convolutional Neural Network)


PROJECT BASED

Name: Suleman
Shivani
Akanksha
Munendra

INTRODUCTION OF CNN:

As a sort of formal definition, “Convolutional Neural Networks or CNNs, are a


special kind of neural network for processing data that has a known, grid-like
topology. Examples include time-series data, which can be thought of as a 1D grid
taking samples at regular time intervals, and image data, which can be thought of
as a 2D grid of pixels. Convolutional networks have been tremendously successful
in practical applications. The name “convolutional neural network” indicates that
the network employs a mathematical operation called convolution. Convolution is
a specialized kind of linear operation.
CNNs are very similar to ordinary Neural Networks — they are made up of
neurons that have learnable weights and biases. Each neuron receives some inputs,
performs a dot product and optionally follows it with a non-linearity. Now for those
who have a general idea on how the neural network works, it will be easy to
visualize. For the others who don’t know anything about Simple Neural Network, it
will be better if you could at least read about the basic working from some resource.

CONVOLUTIONAL LAYER:
The CONVOLUTIONAL LAYER is related to feature extraction. First le us get clear
of the idea of ‘filter’ and ‘convolution’, then, we shall move on to its
implementation in the layer.

Filters: Filters or ‘kernels’ are also an image that depict a particular feature. For
example, let us take the picture of this curve. We take this as a sample that we will
recognize, i.e., determine whether it is present in an image.
Convolution: It is a
special operation applied
on a particular matrix.
The operation involves
multiplying the value of a
cell corresponding to a
particular row and
column, of the image
matrix, with the values of
the corresponding cell in
the filter matrix.

Padding: If we try to visualize the operation of convolution, in our head, as the


filter matrix moves over the whole image, we find
that the no of times, the values of the cells lying
within the matrix is considered for the operation is
more than the no. of times, the values of the cells in
the corners or at the borders, are accounted for. This
implies that the values at the corners or around the
borders are not being given equal weightage. To
overcome this, we add another row and column, of
only 0, at all the sides of the image matrix. This idea
is known as padding. In actual sense, these values
being ‘0’ wouldn’t supply any extra information, but
will help into accounting the previously less-accounted for values to be given more
weightage.

Striding: In ‘strided’ convolution, instead of shifting the


filter one-row or one-column at a time, we shift it,
maybe, 2 or 3 rows or columns, each time. This is
generally done to reduce the no of calculation and also
reduce the size of the output matrix. For large image,
this doesn’t results in loss of data, but reduces
computation cost on a large scale.

PAGE 1
RELU Activation: RELU or Rectified Linear Unit is applied on all the cells of all the
output-matrix. The basic intuition to derive from here is that, after convolution, if a

particular convolution function results in ‘0’ or a negative value, it implies that the
feature is not present there and we denote it by ‘0’, and for all the other cases we
keep the value.
Together with all the operations and the functions applied on the input image, we
form the first part of the Convolutional Block.

POOLING LAYER:
The Pooling layer consist of performing the process of extracting a particular value
from a set of values, usually the max value or the average value of all the values.
This reduces the size of the output matrix. It is common to periodically insert a
Pooling layer in-between successive Conv layers in a ConvNet architecture. Its
function is to progressively reduce the spatial size of the representation to reduce
the amount of parameters and computation in the network, and hence to also
control overfitting. The Pooling Layer operates independently on every depth slice
of the input and resizes it spatially, using the MAX operation. The most common
form is a pooling layer with filters of size 2x2 applied with a stride of 2
downsamples every depth slice in the input by 2 along both width and height,
discarding 75% of the activations. Every MAX operation would in this case be
taking a max over 4 numbers (little 2x2 region in some depth slice). The depth
dimension remains unchanged.

PAGE 2
Together with the CONVOLUTIONAL
LAYER and the POOLING LAYER, we
form the CONVOLUTIONAL BLOCK of
the CNN architecture. Generally, a simple
CNN architecture constitutes of a
minimum of three of these Convolutional
Block, that performs feature extraction at
various levels.

PAGE 3

You might also like