Search

3D segmentation has become a widely used technique. However, automatic segmentation does not deliver high accuracy in optically dense images and manual segmentation lowers the throughput drastically. Therefore, we present a workflow for 3D segmentation being able to forecast segments based on a user-given ground truth. We provide the possibility to correct wrong forecasts and to repeatedly insert ground truth in the process. Our aim is to combine automated and manual segmentation and therefore to improve accuracy by a tunable amount of manual input.

In this paper we will give an overview of our studies on turbulent flow over rough surfaces such as shark skin, which have a drag reducing effect known as shark-skin effect. Our mathematical model is restricted to the flow in the viscous sublayer. Here, the turbulences which occur in the flow above this layer enter through a boundary condition. The flow equations are the steady state Navier-Stokes equations with a Couette flow profile given through two boundary conditions: a diagonal flow on the top and the no-slip condition on the rough surface. Direct simulations are performed via stabilized finite elements. For a better approximation of the boundary isoparametric finite elements are used. Our first calculations give a drag reduction of rough surfaces up to 8% whereas 15% is obtained with an improved model. We will discuss this amount of drag reduction and how it can be compared with experimental results. For gaining insight in the details of the flow, how it is influenced by the different shapes of microstructures, and how their drag reducing mechanism can be explained, we used scientific flow visualization. We will present a short overview of the methods employed.