Page 264 - Ai Book - 10

verifying layer. The process of verifying a signature is commonly resembled with face recognition technique
because a signature contains unique features that make it distinct from the others.

u Image Tagging: Image tagging algorithms are the most basic type of
image classification which is being performed by the CNNs. The image
tag is a word or a word combination that describes the images and
makes it easier to find. Google, Facebook, and Amazon are using this
technique. It is also one of the foundation elements of visual search.
Tagging includes recognition of objects and even sentiment analysis of
the picture tone.

u Droning Industry: Nowadays, drones have been an excellent boon for mankind in mapping regions from an
aerial perspective and executing human activities in conditions that are either too dangerous or difficult.
Convolution neural network , a representative of deep learning technology is often used in the field of
droning industry to recognise images.

DEPLOYING A CNN

As you know, CNNs are a driving force for a
new age of machine learners that enable both
researchers and industries to rethink their
approach in interpreting the world around
them. CNNs, one of the most commonly used
neural networks for performing high scale
data intensive tasks like identification, classification and automation. After learning a short
description about CNNs, let us understand the procedure of deploying a CNN with the help
of the following figure:

u Convolution Layer: The first layer of CNN is Convolution Layer. The high-level features of
an image such as edges would be extracted with the help of Convolution operation. In
general, several kernels are available in the Convolution layer that are used to generate
features from an input image. The featured output or the output of the Convolution
layer is called the Feature map. Sometimes, feature maps are also referred to as

138
138