Scribd
Upload
5 views

IP With Python Libraries

IP with python libraries

Uploaded by

VivekanandaGN
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
5 views

IP With Python Libraries

IP with python libraries

Uploaded by

VivekanandaGN
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 17

IMAGE

PROCESSING

Using Python libraries


Scikit image , openCV
About Me
Hello, my name is Nidhi Maheshwari, and I am a Data Scientist .

I've delved deep into the intricate relationship between image processing and
computer vision. While computer vision aims to interpret and understand visual
data, it heavily relies on the foundational work done by image processing.
Techniques like noise reduction, image enhancement, and segmentation are
essential preprocessing steps that enable accurate face recognition, object
detection, and more.
Project
Overview
This project offers a comprehensive exploration of image
processing, covering both fundamental and advanced
techniques, methodologies, and algorithms. Image
processing often serves as a preprocessing step in
computer vision. For example, before a computer vision
system can recognize faces in an image, image processing
techniques are used to enhance the image quality, remove
noise, or segment the face from the background. So,
successful computer vision often depends on effective
image processing. This project delves into some real-world
applications, providing guidance on the appropriate use of
different Python libraries for specific tasks .
Basic Image
Processing using scikit image

Converting Color Grey Scale Dilation Errosion


Space

Edges Drawing Rectangle


Converting Color Space
google_colab_scikit_image
Original Image in rgb rgb2hsv conversion rgb2yiq conversion
Changing the shape of Image
google_colab_Dilation_Errosion
Dilation Errosion

MORPH_CROSS

MORPH_ELLIPS
Creating Basic Drawing
google_colab_Basic_Drawing_On_Image

Line Square

Naming
Advance Image
Processing using OpenCV

Blending 2 Detecting faces Determining Comparing


Structural Similarity faces
Images
Blending 2 Images
google_colab_Advance__Blending_Two_Images
Input Output
Image1 Blended Image

Input
Image2
Detecting faces
google_colab_haarface_cascade

Input Output
Image Image
Determining Structural Similarity (SIFT)
google_colab_SIFT_ImageComparion_Diff_Scale-Angle

Input
Output
Structure 1
Structural Comparision

Input
Structure 2
Determining Structural Similarity (SIFT)
Output
Input
Structural Comparision
Structure 1

Input
Structure 2
SIFT is effective for comparing architectural structures or similar patterns, but for more advanced tasks
like face identification and comparison, we need to leverage deep learning models such as FaceNet,
which are part of the computer vision domain.
Computer Vision using facenet, deepface, and facerecognization

google_colab_DeepFace
google_colab_Facerecognization
google_colab_facenet.ipynb
Result and Impact
Image processing enhances image quality, enables feature extraction, and
allows for various manipulations. Basic image processing like color conversion,
shape, drawing etc, can be efficiently handled using libraries like scikit-image.
For more advanced image processing tasks blending images, detecting faces
etc, OpenCV provides powerful tools and functions. As the complexity
increases, especially in tasks like face recognition, object detection, and
segmentation, deep learning models and libraries become essential. FaceNet
, deepface, facerecognization are prime example for face recognition, while
TensorFlow and PyTorch are crucial for building and deploying custom deep
learning models. By mastering these image processing techniques and the
appropriate Python libraries, we can pave the way for more successful and
efficient computer vision applications.
Future
Directions
The future of image processing, driven by machine learning, is set to transform various fields.
Deep learning will enhance early cancer detection with advanced imaging, boost real-time
surveillance for better public safety, and advance autonomous driving with precise obstacle
detection. Personalized healthcare will benefit from tailored treatments, while smart cities
will optimize traffic and infrastructure. Advanced robotics will achieve greater precision in
complex tasks. These innovations will revolutionize medical imaging, security, automotive
safety, and beyond, reshaping our world.
Thank You!

Important links :-

google_colab_scikit_image
google_colab_Dilation_Errosion
google_colab_Basic_Drawing_On_Image
google_colab_Advance__Blending_Two_Images
google_colab_haarface_cascade
google_colab_SIFT_ImageComparion_Diff_Scale-Angle.ipynb
google_colab_DeepFace [email protected]
google_colab_Facerecognization
google_colab_facenet.ipynb
linked

You might also like