3-D ISODATA and REGION GROWING technique for segmentation a true color Visible Human dataset

3-D segmentation of true color images and volumes is a big challenger, especially when image Slices contain a noise and a non homogeneity background such as medical volumes. This paper demonstrates a 3-D ISODATA and REGION GROWING technique for segmentation and visualization of Visible Human volumes. The algorithm combines both ISODATA and REGION GROWING technique in three dimensions to produce accurate segmentation results depending on a joint decision based on both color and spatial information Although ISODATA does not have any spatial information, it is applied in 3-D domain to ensure the similarity between all volume´s clusters. The 3-D region growing has been used to determine the region of interest of the volume. The algorithm has been applied for segmentation a Visible Human white matter as a region of interest. ISODATA is not necessary to determine the number of clusters in the beginning so it does not suffer from this problem as K-Means. The performance of the algorithm are studied and compared with other hybrid algorithm results using Scatter Matrices and Separability criteria.