A MR Image Segmentation Method Based on Nonlinear Anisotropic Diffusion Filter

Objective Introduce an effective method for segmenting the brain from head magnetic resonance (MR) images. A multistage process is needed. At first, roughly delineate the brain structure from head MR images as the region of interesting (ROI) manually; next, the ROI is to be processed by a nonlinear anisotropic diffusion filter; and then, the filtered image is used as an image mask after binary process; at last, the brain structure is extracted from the original image using the image mask. The brain structure of head MRI can be segmented by this method effectively. When using image processing techniques based on a nonlinear anisotropic diffusion filter to segment the brain structure in MRI, the segmentation result is correct. It has strong robustness and stability.