A background removing method of MR images and its application in the intensity nonuniformity correction methods

Intensity nonuniformity correction is a necessary preprocessing method in MR images segmentation. Most of intensity nonuniformity correction methods need to remove the background of the images. In this paper, we propose a segmentation method based on the region growing to remove the background as a instead of the threshold method which is wildly used in the intensity nonuniformity correction methods. We tested this method on a lot of MR images and applied it to the N3 method. The experiments showed that it performances better than the threshold method.