PDE based algorithms for smooth watersheds

      Tracking of front propagation for water immersion has many applications. Traditionally, the front is tracked by fast sorting algorithms. In this work, we propose a PDE based method to track the front. The main advantage of using a partial differential equation to track the immersion front is that the method becomes versatile and may easily be stabilized by introducing regularization terms. We demonstrate the advantage of the new method for image segmentation. Coupling the geometric approach with a proper “merging strategy” creates a robust algorithm which minimizes over- and under-segmentation even without predefined markers.