APPLICATION OF SUBBAND ADAPTIVE THRESHOLDING TECHNIQUE WITH NEIGHBOURHOOD PIXEL FILTERING FOR DENOISING MRI IMAGES

摘要

The image de-noising naturally corrupted by noise is a classical problem in the field of signal or image processing. Image denoising has become an essential exercise in medical imaging especially the Magnetic Resonance Imaging (MRI)..We propose a new method for MRI restoration. Because MR magnitude images suffer from a contrast-reducing signal-dependent bias. Also the noise is often assumed to be white, however a widely used acquisition technique to decrease the acquisition time gives rise to correlated noise. Subband adaptive thresholding technique based on wavelet coefficient along with Neighbourhood Pixel Filtering Algorithm (NPFA) for noise suppression of Magnetic Resonance Images (MRI) is presented in this paper. A statistical model is proposed to estimate the noise variance for each coefficient based on the subband using Maximum Likelihood (ML) estimator or a Maximum a Posterior (MAP) estimator. Also this model describes a new method for suppression of noise by fusing the wavelet denoising technique with optimized thresholding function. This is achieved by including a multiplying factor (a) to make the threshold value dependent on decomposition level. By finding Neighbourhood Pixel Difference (NPD) and adding NPFA along with subband thresholding the clarity of the image is improved. The filtered value is generated by minimizing NPD and Weighted Mean Square Error (WMSE) using method of leastsquare.Areduction in noise pixel is well observed on replacing the optimal weight namely NPFA filter solution with the noisy value of the current pixel. Due to this NPFA filter gains the effect of both high pass and low pass filter. Hence the proposed technique yields significantly superior image quality by preserving the edges, producing a better PSNR value. To confirm the efficiency this is further compared with Median filter, Weiner Filter, Subband thresholding technique along with NPFA filter.