Segmentation of MR Brain Images Using Particle Swarm Optimization (PSO) and Differential Evolution (DE)
Magnetic resonance imaging (MRI) is a powerful tool for clinical diagnosis because it allows to distinguish different tissues and allows multiple modalities (T1, T2, ...) each having particular properties. In this work, the segmentation of MR Brain images is considered as an optimization problem and solved using evolutionary algorithms: particle swarm optimization (PSO) and differential evolution (DE),. The process of segmentation is done with multilevel fuzzy thresholding. The performances of three approaches were compared using the fidelity criterion: the peak-to-signal-noise (PSNR) ratio. The methods adopted provide good results in terms of accuracy and robustness. However, the PSO is the most efficient.
Keywords: Optimization, Segmentation, Particle swarm optimization, Differential Evolution Algorithm, MR images.
Download Full-Text
ABOUT THE AUTHOR
Assas Ouarda
Department of Computer Science, Laboratory Analysis of Signals and Systems (LASS) University of M’sila M’sila, Algeria
Assas Ouarda
Department of Computer Science, Laboratory Analysis of Signals and Systems (LASS) University of M’sila M’sila, Algeria
