[Seminar] Image and Shape Analysis of Biological Specimens by Daniel Baum

[Seminar]

Title: Image and Shape Analysis of Biological Specimens

Speaker: Professor. Daniel Baum

Institution: Zuse Institute Berlin, Germany

Abstract:

 Today, image acquisition is usually the first step to analyze the morphology of biological specimens. Depending on the size of the objects of interest, medical CT, micro-CT or even electron microscopy are used. No matter which of these image acquisition techniques is applied, one is challenged with three-dimensional (3D) image data that is often large and complex. However, investigating the image of a single specimen is usually not enough but several or even many specimens need to be considered in order to allow for statistical analysis and comparisons. Hence, fast automatic or at least semi-automatic techniques are required to handle the huge amount of data. Once the 3D image has been segmented and interesting structures have been extracted, geometric and shape analysis techniques are applied in a second step, for example, to compare structures between several individuals, species or genera.

 In my talk, I will present some work on image segmentation and shape analysis that has been carried out in the department of Visual Data Analysis at ZIB over the past years. Among the presented techniques are the well-known watershed segmentation algorithm that can be applied to different kinds of scalar fields such as the original gray values, but also distance maps, and ambient occlusion fields. Further techniques include template matching and model-based segmentation. Finally, I will show some results for shape model-based morphometrics and shape morphing.

Biography:

  Daniel Baum is head of the working group »Image Analysis in Biology and Materials Science« at the Zuse Institute Berlin (ZIB). He studied computer science at the Humboldt-Universität and at the University of Edinburgh. He received his Dr. rer. nat. degree from the Freie Universität Berlin in 2007, specializing in comparative molecular surface analysis. His current research interests are in the fields of molecular visualization, image segmentation and analysis, as well as extraction and shape analysis of biological structures.