Automated background segmentation for Rician noise estimation of noisy MR images

The accurate estimation of Rician noise standard deviation is necessary for effective MR image denoising. In this short paper, we show that background segmentation is desirable for an accurate estimation of Rician noise parameter. Motivated by that observation an automated background segmentation algorithm is developed by combining morphological operations and active contour model in order to get more desired results. A test set MR images on 62 slices of human knee is used for illustration purpose. The proposed method is compared with some existing noise estimation methods and is shown to produce more accurate results.