Automatic segmentation of skin and bone in CT images using iterative thresholding and morphological image processing

In this paper, we propose a fast and efficient method to extract skin and bone automatically. First, we smooth the images by applying anisotropic diffusion filter to remove noise. Then, the whole body is detected by using thresholding which is set automatically. In addition, we segment the contour of the skin by using morphological operators and connected component labeling (CCL). Finally, we extract the bone with iterative thresholding.