A K-Means-Galactic Swarm Optimization-Based Clustering Algorithm with Otsu's Entropy for Brain Tumor Detection

Image segmentation is a technique in order to segment an image into various parts and derive meaningful information out of each one. In this article, problem of image segmentation is applied on brain MRI images. This is done in order to detect and capture the location, size and shape of five differe...

Full description

Saved in:

Bibliographic Details
Published in:	Applied artificial intelligence Vol. 33; no. 2; pp. 152 - 170
Main Authors:	Nanda, Satyasai Jagannath, Gulati, Ishank, Chauhan, Rajat, Modi, Rahul, Dhaked, Uttam
Format:	Journal Article
Language:	English
Published:	Philadelphia Taylor & Francis 28-01-2019 Taylor & Francis Ltd Taylor & Francis Group
Subjects:	Algorithms Brain Cluster analysis Clustering Computer simulation Entropy of solution Fitness Genetic algorithms Image detection Image segmentation Magnetic resonance imaging Medical imaging Optimization Tumors Vector quantization
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Image segmentation is a technique in order to segment an image into various parts and derive meaningful information out of each one. In this article, problem of image segmentation is applied on brain MRI images. This is done in order to detect and capture the location, size and shape of five different types of tumors. Here, image segmentation is viewed as an clustering problem and a new hybrid K-means Galatic Swarm Optimization (GSO) algorithm is proposed for effective solution. The Otsus entropy measure is used as the fitness function for deriving the segments. Extensive simulation studies with five performance measures on five different brain MRI images reveal the superior performance of the proposed approach over GSO, Real Coded Genetic Algorithm (RCGA), and K-Means clustering algorithms.
ISSN:	0883-9514 1087-6545
DOI:	10.1080/08839514.2018.1530869