Medical image segmentation based on watershed and graph theory

Strong noise, poor gray-scale contrast, blurred margins of tissue are characteristics of medical images. Extracting object of interest in medical images is challenging. A segmentation approach that combines watershed algorithm with graph theory is proposed in this paper. This algorithm reconstructs gradient before watershed segmentation, based on the reconstruction, a floating-point active-image is introduced as the reference image of watershed transform. Finally, a graph theory based algorithm Grab Cut is used for fine segmentation. False contours of over-segmentation are effectively excluded and total segmentation quality significant improved as suitable for medical image segmentation.