Scribd
Upload
25 views

DIP Unit-3 Chapter-2 Lecture 4

Uploaded by

Vivek Dangi
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
25 views

DIP Unit-3 Chapter-2 Lecture 4

Uploaded by

Vivek Dangi
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

Apex Institute of Technology

Department of Computer Science & Engineering


Bachelor of Engineering (Computer Science & Engineering)
Digital Image Processing– (20CST-481)
Prepared By: Mr. Aadi Partap Singh (E15043)

DISCOVER . LEARN . EMPOWER


06/22/2024 1
DIGITAL IMAGE PROCESSING
Course Objective:
During the course, students will be able :

CO
Title Level
Number Will be covered in
To Understand the fundamental of digital Understand this lecture
CO1 image processing with python.
To acquire the knowledge to apply various
CO2 Understand
image processing techniques and tools.
To learn the practical applications of
CO3 image processing steps to real world Understand
problem.

06/22/2024 2
DIGITAL IMAGE PROCESSING
Course Outcome:
Upon successful completion of this course, students will be able to:

CO
Title Level
Number
Will be covered in
Understand this lecture
CO1 Fundaments and techniques implemented in
digital image processing technologies

Understanding the various filters applications,


Understand
CO2 smoothing applications and techniques by
image processing implementation by python
Acquiring knowledge on various compression
CO3 and segmentation techniques, for image Understand
enhancement methods.

06/22/2024 3
Welcome to the session of

Unit-3:Application of IP in Machine learning

Chapter-2: Image processing using machine


learning

06/22/2024 4
INTRODUCTION

• Nowadays, computers can also describe the various elements in pictures and write
short sentences describing each segment with proper English grammar apart from
just classifying photos.

• Computers does this with the help of Deep Learning network (CNN) which actually
learns patterns that naturally occur in photos.

• Imagenet is one of the biggest databases of labeled images to train the


Convolutional Neural Networks using GPU-accelerated deep learning frameworks
such as Caffe2, Chainer, Microsoft Cognitive Toolkit, MXNet, PaddlePaddle,
Pytorch, TensorFlow, and inference optimizers such as TensorRT.

https://www.infrrd.ai
2
DEEP LEARNING

• Deep learning, also called neural networks, is a subset of machine learning that
uses a model of computing that’s very much inspired by the structure of the
brain.

• It is already working in Google search and in image search. It allows you to


image-search a term like ‘hug.’

• It’s used to getting you Smart Replies to your Gmail. It’s in speech and vision.
DEEP LEARNING NEURAL NETWORK

• Deep Learning models, with their multi-level


structures, as shown beside, are very helpful in
extracting complicated information from input
images.

• Convolutional neural networks are also able to


drastically reduce computation time by taking
advantage of GPU for computation which many
networks fail to utilize. Deep learning neural network (CNN)
IMAGE DATA PREPARATION
USING DEEP LEARNING

Image
Data Labelling Using RCNN
Classification
IMAGE
CLASSIFICATION
• For increased accuracy, Image classification
using
CNN is most effective.

• Let us take an example of beauty and


pharmacy
products, as our initial training data set.

• The most common image data input parameters are


the number of images, image dimensions, number of
channels, and number of levels per pixel.
• With classification, we get to categorize images (in
this case, as beauty and pharmacy). Each category Image classification using CNN
again has different classes of objects as shown in the
picture.
DATA LABELING
• It’s better to manually label the input data so that the deep learning algorithm can eventually learn to
make the predictions on its own.

• Some off the shelf manual data labelling tools are given here. The objective at this point will be mainly
to identify the actual object or text in a particular image, demarcating whether the word or object is
oriented improperly, and identifying whether the script (if present) is in English or other languages.

• To automate the tagging and annotation of images, NLP pipelines can be applied. ReLU (Rectified
Linear Unit) is then used for the non-linear activation functions, as they perform better and decrease
training time.

• To increase the training dataset, we can also try data augmentation by emulating the existing images and
transforming them.

• We could transform the available images by making them smaller, blowing them up, cropping elements
etc.
USING RCNN
• With the usage of region-based convolution neural network aka RCNN, locations
of
objects in an image can be detected with ease.

• Within just 3 years the R-CNN has moved from Fast RCNN, Faster RCNN to Mask
RCNN, making tremendous progress towards human-level cognition of images.

• In the below slide, we have shared an example of the final output of the image recognition
model where it was trained by deep learning CNN to identify categories and products in
images.

11
IMAGE RECOGNITION OUTPUT USING
RCNN

Category Detection Product Detection

12
Assessment Pattern
S.No. Item Number/semester Marks
1 MSTs 2 20 per each
2 Quiz 2 per unit 4 per each quiz
3 Time bound surprise test 3 (one per unit) 12 per each test

4 Assignments 3 (one per unit) 10 per each Assignment

Engagement task (non


5 One per each topic depends
gradable)
Attendance +
6 Above 90% 2
Engagement score
Internal (division as mentioned above points 1-6) 40

External 60

Total 100

06/22/2024 13
THANK YOU

For queries
Email: [email protected]

06/22/2024 14

You might also like