Anagha-Sankar/Bayes-Classifier

Classification of image pixel into 2 classes using Baye's Decision Rule

BAYE'S CLASSIFIER

CONTENTS:

• Description

• Requirement

• Installation

• Directory Structure

• Usage

• Credit

  Description

  Bayesian Classifier uses the Bayes Theorem. Bayes Classifier is one of the simple and most effective Classification algorithms. It is a probabilistic classifier, which means it predicts on the basis of the probability of an object.

  Requirement

  1. Python 3
  2. Jupyter Notebook
  3. Numpy
  4. Opencv
  5. PIL

  Installation

  1. Python

     • Visit and download Python from https://www.python.org/downloads/

     • Install and add Python to path

       python3 -V
       

       Install pip

       sudo apt install python3-pip
       

  2. Numpy

     Using pip,

     pip install numpy
     

  3. Opencv

     In command line, change directory to where pip is present

     pip install opencv-python
     

  4. PIL

     In command line or powershell

     python3 -m pip install --upgrade Pillow
     

  5. Jupyter notebook

     Install the classic Jupyter Notebook using:

     pip install notebook
     

     To run the notebook

     jupyter notebook
     

  Directory Structure

  .
  ├── src                     	# Source files
  │   ├── Bayes Classifier.ipynb	# Jupyter Notebook
  |	├──band1.jpg	# Input image 1
  |	├──band2.jpg	# Input image 2
  |	├──band3.jpg	# Input image 3
  |	├──band4.jpg	# Input image 4
  |	├── River		# 50 Coordinates of River class
  |	├── NonRiver	# 100 Coordinates of Non-River class
  |	├── Output_Images
  |		├── river03non07.jpeg
  |		├── river03non07.jpeg
  |		├── river03non07.jpeg
  ├── REPORT.md
  └── README.md
  

  Usage

  The source code (jupyter notebook) is present in the 'src' folder. The 4 input band images(R,G,B,I) are band1.jpg, band2.jpg, band3.jpg and band4.jpg. Coordinates of river and non-river for training pixel are in the files 'River' and 'NonRiver'. In the 'Output_Images' folder, the three output images for different Prior Probabilities are included.

  Project.md explains the source code.

Credit

To record coordinates, https://www.mobilefish.com/services/record_mouse_coordinates/record_mouse_coordinates.php