Robert C. Acar

We don’t have enough information about this author to calculate their statistics. If you think this is an error let us know.
Learn More
Several models for image segmentation can be cast in variational form, meaning that they amount to minimising a certain “energy”. Direct iterative methods are bound to fail when the energy functional is not convex. We discuss the principles and implementation of some algorithms which descend from Blake and Zisserman’s “graduated nonconvexity algorithm”, and(More)
Variational methods for image denoising consist of minimizing a functional which incorporates both the data and some penalty term. Choosing the penalty term to involve the total variation of the image has the advantage of cleaning speckles without smoothing out the edges. Our goal is to investigate the use of genetic algorithms to minimize the functional.
  • 1