A new method to remove salt & pepper noise in Magnetic Resonance Images

Magnetic Resonance Images (MRI) are corrupted by salt and pepper noise mainly due to sensor faults of image acquisition devices, sharp and sudden disturbance in image signal which in turn degrades the image quality. For removing salt and pepper noise from corrupted MR images researchers are using so many algorithms. In this paper, a proposed method based a median filter is introduced to remove salt and pepper noise. The proposed algorithm is to detect the corrupted pixel first since the impulse noise only affect certain pixels in the image and the remaining pixels are uncorrupted. The corrupted pixels are replaced by an estimated value using the proposed median filter. Comparison of the proposed method with traditional median filter is provided. The proposed median filter and standard median filter algorithms will be implemented by using MATLAB R2015a. The peak signal-to-noise ratio (PSNR) will be used to quantify the amount of noise of the MR images.