Image-Processing-IMG-IMPACT

In the project IMG-IMPACT, several Image Processing Techniques are used, to IMPACT THE IMAGES.

This project is based on the Linear Algebra project, which was part of the course UE20MA251.

Table of Contents

Introduction

Prerequisites and Techstack

Steps for Execution

Screenshots

Usage

Skip to END...

Introduction

IMG-IMPACT is a project on Impacting the Images , using various Image Processing Technologies and Techniques .

These Techniques include various Filters like Median filter, Gaussian filter, Linear filter, etc...

The other Technologies implemented here are Scaling, Cropping, Rotation, etc...

Files :

In the IMG-IMPACT Directory there are 4 directories:

• 1- Images Directory- contains all the Images used in the programs.

• 2- Individual_Filters&Techniques Directory- contains the techniques and filters implemented separately.

• 3- All_Filters_Combined Directory- contains one program for the techniques and filters combined.

• 4- Extra_Contents Directory- contains extra information about the techniques and filters used.

Repository Structure :

IMG-IMPACT repo structure click...

Below is the structure of the IMG-IMPACT project repository

  Image-Processing-IMG-IMPACT/
  │   
  ├── IMG-IMPACT/           # Project Folder
  │   │              
  │   ├── All_Filters_Combined/              # Folder3
  │   │    └── IMG-IMPACT.py
  │   │     
  │   ├── Extra_Contents/                    # Folder4
  │   │    ├── IMG-IMPACT_Slides.pdf
  │   │    └── IMG-IMPACT_Information.pdf
  │   │ 
  │   ├── Images/                            # Folder1
  │   │    ├── 1_2_rover.jpg
  │   │    ├── 3_4_turtle.jpg
  │   │    ├── 4_CroppedImage.jpg
  │   │    ├── 5_desktop.jpg
  │   │    ├── 6_scenary.jpg
  │   │    ├── 7_8_emoji.png
  │   │    ├── 9_person.png
  │   │    ├── 10_pig.jpg
  │   │    ├── 11_apple.jpg
  │   │    ├── <no.>_new_<any name>.jpg      
  │   │          # These files are the resultant images after running the files, present in the 'Individual_Filters_Techniques' Folder.
  │   │    ├── <no.>_fnew_<any name>.jpg      
  │   │          # These files are the resultant images after running the file, present in the 'All_Filters_Combined' Folder.
  │   │    └── IMG-IMPACT.png    #logo
  │   │ 
  │   ├── Individual_Filters_Techniques/     # Folder2
  │   │    ├── 1_ColourFilter.py
  │   │    ├── 2_GrayScale.py
  │   │    ├── 3_ImageReconstruction.py
  │   │    ├── 4_ImageCropping.py
  │   │    ├── 5_LinearFilter.py
  │   │    ├── 6_ImageResizing.py
  │   │    ├── 7_ImageRotation.py
  │   │    ├── 8_GaussianFilter.py
  │   │    ├── 9_MedianFilter.py
  │   │    ├── 10_ImageScaling.py
  │   │    └── 11_ImageShearing.py
  │   │ 
  │   └── IMG-IMPACT.png   # Project Logo
  │   
  └─── README.md           # Repository README
  

Underlying concepts / theory :

1. Inclusion of Image Processing Techniques

   The project IMG-IMPACT involves the implementation of Image Processing Techniques.

2. Manipulation of Images and Extraction of Insights

   Image processing entails the manipulation of imagesto extract useful insights from them.

3. Representation of Images as Grids of Pixels

   Images can be represented as a grid of n x n small pieces, which are called pixels.

4. Numerical Representation of Image Pixels

   If we can assign numbers to each colour, then, the grid of pixels can be represented as a numerical matrix.

Back to TOP

Prerequisites and Techstack

• Language :

  Python

• Libraries :

  • OpenCV or cv2
  • Pillow or PIL
  • NumPy
  • argparse
  • matplotlib
  • math

• Concepts :

  Linear transformation, SVD, Fourier transformations (click for more details...)

  • Linear transformation -

    Linear transformations are mathematical operations that maintain the fundamental properties of vector spaces. They are often represented using matrices and are essential in geometry, and transformations in computer graphics. They provide a structured way to describe and manipulate complex relationships between vectors.

  • SVD-

    Singular Value Decomposition, or SVD, is a powerful matrix factorization technique that breaks down a matrix into three simpler components: U, Σ, and V. Here, U and V are orthogonal matrices, and Σ is a diagonal matrix containing singular values. SVD is employed in image compression, as it uncovers hidden patterns and relationships in data

  • Fourier transformations -

    Fourier transformations are mathematical methods used to represent functions or signals in the frequency domain. The Fourier transform allows us to express complex signals as combinations of simpler sinusoidal components, revealing their frequency and amplitude characteristics. This is particularly useful for analyzing and processing signals in fields of image analysis, and data compression.

Back to TOP

Steps for Execution

1. Clone the 'Image-Processing-IMG-IMPACT' github repository.

   git clone https://github.com/ankitacoder3/Image-Processing-IMG-IMPACT.git

   
   

2. Navigate to the 'IMG-IMPACT' Directory in that.

   cd Image-Processing--IMG-IMPACT
   cd IMG-IMPACT

3. Navigate to ANY of the Directories and execute the respective files present in that directory.

   
   

   i. Navigation :
   
   a. Navigate to 'Individual_Filters_Techniques' Directory to execute techniques and filters separately.

   cd Individual_Filters_Techniques

   
   

   b. Navigate to 'All_Filters_Combined' Directory to execute techniques and filters combined.

   cd All_Filters_Combined

   
   

   c. Navigate to 'Extra_Contents' Directory to find extra information about the filters and techniques.

   cd Extra_Contents

   
   

   d. Navigate to 'Images' Directory to edit the images used.

   cd Images

   Back to TOP

   
   

   ii. Execution :

   By opening any file in python idle or command prompt, and running it.

   
   

   a. To execute any file from command prompt, type

   python <filename>

   Then follow the instructions, which appear on command prompt.

   
   

   b. For any file from 'Individual_Filters_Techniques' Directory, say 1_ColourFilter.py, type

   python 1_ColourFilter.py

   Then follow the instructions, which appear on command prompt.

   
   

   c. For any file from 'All_Filters_Combined' Directory, say IMG-IMPACT.py, type

   python IMG-IMPACT.py

   Then follow the instructions, which appear on command prompt.

   
   
   

Note- For some filters and techniques information may not be complete or it may not exist in the'Extra_Contents' Directory.

Back to TOP

Screenshots

• All images before and after using IMG-IMPACT

• Select any filter or technique of your choice

Back to TOP

• Select any image of your choice

• Choose if you want to save editted images or not

• Choose if you want to name the filtered images

• & Much More...

Back to TOP

Usage

• IMG-IMPACT application can transform the pictures, given as input, into different variations of the same picture.

• IMG-IMPACT application has many default images in the 'Images' directory. The user can select any image.

• IMG-IMPACT application has many filters like linear filter, median filter etc. The user can select any filter.

• IMG-IMPACT application also has many operations and methods such as rotating, cropping the image, etc. The user can select any method.

• IMG-IMPACT can be used to create new images, from the old images.

• Finally, IMG-IMPACT can be used to save the new created, editted and filtered images, with default image names or user defined names, as per the user's choice.

• This project could also be used as a project for Linear Algebra courses, like UE20MA251 or ue20ma251.

• Other Examples or Other Usage:

  • Use IMG-IMPACT to batch process a series of images with a specific filter.
  • Employ IMG-IMPACT in a data preprocessing pipeline for machine learning models that require image data.
  • Utilize IMG-IMPACT for educational purposes to demonstrate the effects of different image processing techniques.

  Back to TOP

  
  

Thank you for exploring the IMG-IMPACT project. Happy Image Editting, Filtering and Processing! 🖼️ 🔄