Learning and Classifying Anisotropic 3D morphologies and Structures in Biology and Medicine
Speaker: Hans Burkhardt, Computer Science Department Albert-Ludwigs-Universitaet Freiburg
When: 2007-11-26 10:00:00
Venue: 78-420
Host: Prof Brian Lovell
Abstract:In many pattern recognition problems images have to be classified
independent of their current position and orientation, which is just a
nuisance parameter. Instead of comparing a measured pattern in all possible
locations against the prototypes it is much more attractive to extract
position-invariant and intrinsic features and to classify the objects in the
feature space. Mathematically speaking, patterns form an equivalence class
with respect to a geometric coordinate transform describing motion.
Invariant transforms are able to map such equivalence classes into one point
of an appropriate feature space.
The talk will describe new results for this classical problem and outlines
general principles for the extraction of invariant features from images
(Haar integrals, Lie-Theory, Normalization techniques). The nonlinear
transforms are able to map the object space of image representation into a
canonical frame with invariants and geometrical parameters. Beside the
mathematical definition the talk will concentrate on characterizing the
properties of the nonlinear mappings with respect to completeness and
possible ambiguities, disturbance behaviour and computational complexity. We
especially investigated Haar integrals for the extraction of invariants
based on monomial and relational kernel functions.
Examples and applications will be given from the following projects:
1. Automatic Classification of Airborne Pollen-Grains recorded with a
Confocal Laser Scanning Microscope.
2. Self-Learning Segmentation and Classification of Tissue Cell-Nuclei in
3D Volumetric Data using Voxel-Wise Gray Scale Invariants
3. Development of a fast Search Engine for Protein Fold Databases based on
Invariant 3D Features
4. Automatic Classification of 3D Chromosome Territories of human
lymphocyte Nuclei from confocally scanned two or three color FISH data.
5. Automatic Segmentation of Dendritic Spines in 3D LSM Data
These research results are part of the Freiburg’s life and engineering
initiative where a group of scientists won a competition to establish a new
centre of biological signalling science (BIOSS) at the University of
Freiburg. The Cluster “Centre of Biological Signalling Studies (bioss)”
funded by the initiative of excellence of the German Federal Government and
States will be sponsored with up to € 32,5 million within the next 5 years:
http://www.pr.uni-freiburg.de/pm/2007/pm.2007-10-19.353/
Biography:(biography unavailable)
Type: SAS
Contact:Prof Brian Lovell, seminar host (lovell@itee.uq.edu.au)
or Guido Governatori (ITEE seminar co-ordinator)
(guido@itee.uq.edu.au)