Image segmentation based on the Mumford-Shah model and its variations

Image segmentation is an important step in medical image analysis. The Mumford-Shah (MS) model is a powerful and robust segmentation technique. However, the numerical method of solving the MS model is difficult to implement. Although some alternative approaches have been presented, these methods are either inefficient or applicable only to some special cases. We present a new image segmentation model, which can segment images with different image intensity distributions efficiently. Another difficulty with the MS model is that the segmentation result depends on the initial condition. Our hierarchical segmentation approach can remedy this problem. With the new approach, we can automatically segment complicated medical images and produce good segmentation results.