Noise suppression in brain magnetic resonance imaging based on non-local means filter and fuzzy cluster

Combining Non-Local Means (NLM) filter with appropriate fuzzy cluster criterion, denoising in synthetic brain magnetic resonance imaging (MRI) are evaluated. Experimental results show that noise is effectively suppressed while image details are well kept, compared with the traditional NLM method and wavelet method. Quantitative and qualitative results indicate that the continuous edge and detailed structure are well preserved, artifacts are greatly reduced and brain MR images are typically enhanced. In addition, the computational time is reduced greatly.