A Brain MR Image Segmentation Method Based on Gaussian Model and Markov Random Field

The traditional clustering segmentation method for human brain MRI usually divides pixels by the degree of similarity of the image gray value, but the effect is not ideal for the strong noise or the edge blur brain MRI. So we introduce the FCM membership function into Markov random field, and to cluster by the Markov random field advantage of space correlation, so as to reduce the influence of noise on the results. According to the view point of statistics, we establish a two-dimensional histogram of the human brain MRI to further reduce the impact of noise on image segmentation. Then the two-dimensional histogram is projected into the optimal one-dimensional histogram, so as to diminish the calculation. And the optimal segmentation threshold is obtained depending on fitting of the statistical results by the Gaussian model. In this paper, a large number of experimental results have shown that the method proposed has good segmentation effects for human brain MRI.