Invariance via Group-Integration: A Feature Framework for 3D Biomedical Image Analysis

J. Fehr, O. Ronneberger, J. Schulz, T. Schmidt, M. Reisert, and H. Burkhardt (Germany)


Invariant Features, Group-Integration, Classification, Seg mentation, Landmark-Detection.


One very generic approach towards the construction of features achieves invariance against a certain transfor mation via integration over the respective mathematical group. In this paper we present a general framework for invariant feature design via group integration for biomed ical image analysis on 3D volumetric data, and show the common mathematical context of several previously published invariant methods which are all covered by this basic framework. We focus on the mathematical design paradigms of such features and provide fast implementa tion methods. Further we embed a priori knowledge into the design of highly specialized features. Practical appli cations to these methods are shown on several different datasets with a wide range of different biomedical image analysis problems.

Important Links:

Go Back