Texture Feature based Automated Seeded Region Growing in Abdominal MRI Segmentation

A new texture feature based seeded region growing algorithm is proposed for the automated segmentation of organs in Abdominal MR image. Co-occurrence texture feature and semi-variogram texture feature are extracted from the image and the seeded region growing algorithm is run on these feature spaces. With a given Region of Interest(ROI), a seed point is automatically picked up based on three homogeneity criteria. A threshold is then obtained by taking a lower value just before the one causing ´explosion´. This algorithm is tested on 12 series of 3D abdominal MR images.