Substructure Matching with Frame Features

Applications in Medical imaging and Molecular Biology

[ Xavier Pennec ]

Browser configuration

Most approaches in image processing rely on the extraction of features from the image that are supposed to contain most of the geometric information. However, due to the difficulty to handle generic geometric features, one usually only consider point features.

In this demo we show how that frame features naturally arise when modeling the 3D structure of proteins or medical images and how to use properly these features in a substructure matching algorithm. We present examples of applications for 3D substructure matching of proteins and medical images registration.

Modeling data by frames

Medical imaging:

Molecular biology:

Substructure matching with geometric hashing


Molecular biology:

Medical imaging:

Conclusion and references

[ Xavier Pennec Home Page ]