An improved biogeography based optimization approach for segmentation of human head CT-scan images employing fuzzy entropy

  • Authors:
  • Amitava Chatterjee;Patrick Siarry;Amir Nakib;Raphael Blanc

  • Affiliations:
  • Laboratoire Images, Signaux et Systèmes Intelligents (LiSSi, EA 3956), Université Paris XII Val de Marne, 61 avenue du Général de Gaulle, 94010 Créteil, France and Jadavpu ...;Laboratoire Images, Signaux et Systèmes Intelligents (LiSSi, EA 3956), Université Paris XII Val de Marne, 61 avenue du Général de Gaulle, 94010 Créteil, France;Laboratoire Images, Signaux et Systèmes Intelligents (LiSSi, EA 3956), Université Paris XII Val de Marne, 61 avenue du Général de Gaulle, 94010 Créteil, France;Fondation Adolphe De Rothschild, Département Neurosciences, Service de Neuroradiologie Interventionnelle, 25 rue Manin, 75019 Paris, France

  • Venue:
  • Engineering Applications of Artificial Intelligence
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

The present paper proposes the development of a three-level thresholding based image segmentation technique for real images obtained from CT scanning of a human head. The proposed method utilizes maximization of fuzzy entropy to determine the optimal thresholds. The optimization problem is solved by employing a very recently proposed population-based optimization technique, called biogeography based optimization (BBO) technique. In this work we have proposed some improvements over the basic BBO technique to implement nonlinear variation of immigration rate and emigration rate with number of species in a habitat. The proposed improved BBO based algorithm and the basic BBO algorithm are implemented for segmentation of fifteen real CT image slices. The results show that the proposed improved BBO variants could perform better than the basic BBO technique as well as genetic algorithm (GA) and particle swarm optimization (PSO) based segmentation of the same images using the principle of maximization of fuzzy entropy.