Wavelet domain filtering of MR image sequences with appropriate filtering approach based on MR image type

Magnetic Resonance Imaging (MRI) is becoming popular in medical diagnosis due to its diagnostic applications and advantages over X- rays. But diagnosis task becomes difficult when noise gets introduced in MR images. Also now a days trend is leaning towards automatic diagnosis. Hence in preprocessing task de-noising has become a challenge. Denoising methods based on linear filters cannot preserve image structures such as edges in the same way that methods based on nonlinear filters can do it. In this paper, an algorithm is introduced that uses wavelet based multiresolution analysis and adaptive filtering which can effectively remove noise from image data. Here, we have used Discrete wavelet transform (DWT) and Un-decimated wavelet transform (UDWT), which are functionally somewhat different, for de-noising of MR images and compared the results of the two. Also we have compared different wavelet families and filters and suggested the best combination for de-noising technique.