Summary: A NEURAL NETWORK FOR
CALCULATING ADAPTIVE SHIFT AND ROTATION
INVARIANT IMAGE FEATURES
Sabine Kr¨oner
Technische Informatik I
Technische Universit¨at Hamburg­Harburg
21071 Hamburg, Germany
Tel/Fax: +49 [40] 7718 2539 / 7718 2911
e­mail: kroener@tu­harburg.d400.de
ABSTRACT
Shift and rotation invariant pattern recognition is usu­
ally performed by first extracting invariant features from
the images and second classifying them. This poses the
problem of not only finding suitable features but also a
suitable classifier.
Here a structured invariant neural network architec­
ture (SINN) is presented that performs adaptive in­
variant feature extraction and classification simultane­
ously. The network is sparsely connected and uses
shared weight vectors. As a result features especially

Source: Albert-Ludwigs-Universität Freiburg, Institut für Informatik,, Lehrstuhls für Mustererkennung und Bildverarbeitung

Collections: Computer Technologies and Information Sciences