FCM Based Automatic Thresholding Algorithm to Segment the Brain MR Image

Segmenting the various brain tissues is very valuable in the analysis of brain MR image data, and has a wide range of applications involving clinical analysis, visual inspection etc., it is a challenging due to poor image contrast and artifacts that result in missing or diffuse organ/tissue boundaries. A simple automatic thresholding algorithm based on FCM for segmentation of brain MR image is developed. The thresholds, determined by the information of the result of FCM and Otsu's thresholding method, is used to extract cerebrum region (interesting region), and segment it by FCM. This paper performs the uniformity measure to the segmented image. The results show that the quality of interesting region of segmented images is satisfied. The proposed segmentation algorithm can not only overcome the noise of image, but also can extract the cerebrum region and segment.