Edge-based Segmentation Using Robust Evolutionary Algorithm Applied to Medical Images

  • Authors:
  • Yamina Mohamed Ben Ali

  • Affiliations:
  • Research Group on Artificial Intelligence, Computer Science Department, Badji Mokhtar University, Annaba, Algeria BP 12

  • Venue:
  • Journal of Signal Processing Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Although medical image segmentation is a hard task in image processing, it is possible to reduce its complexity by considering it as an optimization problem. This paper presents a robust evolutionary algorithm based on a cost minimization function to segment and to extract image edges. Since, the goal is to outperform a high edge detection quasi independent from the input problem characteristics, an adaptive detector is considered. As a first step, the main evolutionary algorithm parameters are highlighted based on an adaptive parameterization to overcome convergence problem. In a second stage, the reached optimal setting is applied on medical images to exhibit the quality of the proposed algorithm.