1. CONVOLUTIONAL NEURAL NETWORKS
CNN stands for Convolutional Neural Network, which is a type of
deep neural network that is primarily used for image and video
recognition and classification tasks.
CNN consists of three main layers:
• Convolutional layers
• Pooling layers
• Fully Connected Layers
2. 1.CONVOLUTIONAL LAYERS
Kernel: A kernel is typically a square matrix, such as 3x3 or 5x5.That performs a
convolution operation on the input image. The kernel is applied to each region of the
input image, and the resulting values are summed to produce a single output value.
Stride: Stride refers to the number of pixels that the kernel is shifted when performing
the convolution operation on the input image.
Padding :Padding is a technique used to preserve the spatial dimensions of the input
image when performing the convolution operation.
3. 5 x 5
3 x 3
3 x 3
(n-f+1)
(n)
(f)
If there is stride then (n-f)/s +1
4.
5. 2.POOLING LAYERS
Pooling layers Pooling is the process of down-sampling the
output of a convolutional layer by taking the maximum or
average value in a specific region of the feature map.
There are two main types of pooling:
Max Pooling : In max pooling, It selects the maximum value
within each window.
Average Pooling: In average pooling, It selects the Average
value within each window.
6. 3.FULLY CONNECTED
LAYERS
It is also called as
dense layer .It is a type
of layer in a neural
network where every
neuron in the layer is
connected to every
neuron in the previous
layer.