An adaptive denoising method used in MRI
Huaizhong Zhang, Xianghua Xie · 2014
This paper proposes an adaptive denoising method that can significantly reduce Rician noise in magnetic resonance imaging (MRI). We use a Rayleigh kernel in the denoising processing of self-snakes instead of the Gaussian kernel that is usually used. The Rayleigh kernel is adaptively constructed according to the estimated standard deviation of Rician noise in images. The numerical implementation is carried out by applying the level-set techniques with a semi-implicit scheme. Experimental results in both synthetic and real images demonstrate the effectiveness and advantages of the proposed method in comparison with the traditional methods.