Segmentation of brain tissue based on connected component labeling and mathematic morphology

In order to realize more accurate and efficient segmentation of the Visible Human dataset, an indirect algorithm based on connected component labeling and mathematic morphology was proposed for brain tissue segmentation in this paper. Initially, the region of nonbrain tissue was roughly distinguished through connected component labeling. Then its edge was refined by means of dilation and erosion to complete the segmentation of nonbrain tissue. Finally, extraction of brain tissue was realized by eliminating the segmented nonbrain tissue from the original image. The experimental results show that the proposed algorithm can lead to satisfactory segmentation of brain tissue.