earticle

영어

Two region graph cut image partitioning after mapping entire image of brain in brain MRI. It is seen that piecewise constant model of the graph cut formulation becomes applicable when the image data is transformed by a main function. The proposed function contains data term to evaluate the deviation of the transformed data within two segmentation region, from the piecewise constant model, and a smoothness boundary preserving regularization term. Using a common function, energy minimization typically consists of iterating image partitioning by graph cut iterations and evaluations of region parameters via fixed point computation. The method results in good segmentations and runs faster the graph cut methods. The segmentation of MRI image is challenging due to the high diversity in appearance of tissue among the patient. An automatic interactive brain MRI image segmentation system with the ability to adjust operator control is achieved through this method.