Genetic Programming Based Composite Filter for Rician Noise Reduction

Composite filters based on Mathematical Morphological (MM) operators are getting considerable attraction in denoising Magnetic Resonance (MR) images. However, most of the approaches depend on pre-fixed combination of MM operators. In this paper, we propose a genetic programming (GP) based approach for denoising MR images. An Optimal Composite Morphological Supervised Filter FOCMSF is developed through a certain number of generations by combining the gray-scale 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 the standard MRI data sets and its performance is compared with previously proposed state-of-the art methods. Comparative analysis demonstrates the superiority of the proposed GP based scheme over the existing approaches.