Muhammad SHARIF¹, Muhammad Arfan JAFFAR¹, and Muhammad Tariq MAHMOOD^2,*

¹Department of Computer Science, National University of Computer & Emerging Sciences (FAST-NU), Islamabad, Pakistan
²School of Computer Science and Engineering, Korea University of Technology and Education, Cheonan 330-708, Korea

(Received March 18, 2013; Accepted April 30, 2013)

We propose a genetic programming (GP)-based approach for noise reduction from magnetic resonance imaging (MRI). An optimal composite morphological supervised filter (F_OCMSF) is developed through a certain number of generations by combining gray-scale mathematical morphological (MM) operators under a fitness criterion. The proposed method does not need any prior information about the noise variance. The improved performance of the developed filter is investigated using simulated and real MRI datasets. Comparative analysis demonstrates the superiority of the proposed GP-based scheme over the existing approaches.

Key words: Rician noise, genetic programming, mathematical morphology, image denoising