Segmentation of Images Using K-Means Clustering Algorithm

EasyChair Preprint 6018

Abstract

In a decision-oriented application, image segmentation is one of the most commonly used methods for correctly classifying the pixels of an image. The process of partitioning an image into multiple segments is known as Image segmentation. An image is divided into distinct regions with high similarity between pixels in each region and high contrast between regions. Threshold based, edge based, cluster based, neural network based are some of the techniques used for image segmentation. From these different techniques one of the most efficient methods is the clustering method. K-means clustering, Fuzzy C-means clustering, mountain clustering method, and subtractive clustering method are all methods used for image segmentation using clustering. Here, in this project we use one of the most efficient method called K-means clustering algorithm to find a segmented image. Here we take number of clusters as input and image segmentation is done based on it. Image segmentation has become a popular technique in the medical field, where it is used to isolate a region of interest from a background image.

Keyphrases: K-means clustering algorithm, Segmented image, image segmentation

@booklet{EasyChair:6018,
  author    = {Durganala Srenitha},
  title     = {Segmentation of Images Using K-Means Clustering Algorithm},
  howpublished = {EasyChair Preprint 6018},
  year      = {EasyChair, 2021}}