Self-Organized Feature Extraction Achieved with a Parameterized Filterbank

Authors:
Temujin Gautama;Marc M. Van Hulle
Affiliations:
Laboratorium voor Neuro- en Psychofysiologie, K.U. Leuven, Campus Gasthuisberg, Herestraat 49, B--3000, Leuven, Belgium. E-mail: marc@neuro.kuleuven.ac.be;Laboratorium voor Neuro- en Psychofysiologie, K.U. Leuven, Campus Gasthuisberg, Herestraat 49, B--3000, Leuven, Belgium. E-mail: marc@neuro.kuleuven.ac.be
Venue:
Neural Processing Letters
Year:
1999

Citing 3
Cited 1

What is the goal of sensory coding?

Neural Computation
Self-organized formation of various invariant-feature filters in the adaptive-subspace SOM

Neural Computation
Optimally adaptive transform coding

IEEE Transactions on Image Processing

Reverse engineering expert visual observations: From fixations to the learning of spatial filters with a neural-gas algorithm

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Filterbanks are often used as preprocessing stages in many applicationsand can be used for e.g. feature extraction or dimensionality reduction. In this paper, an unsupervised learning algorithm is described that develops a bank of parameterized (Gabor) filters, starting from a training set. It forces the filters to partition the input space in an equitable manner: each filter is tuned to a different frequency region and contributes equally to the extraction of localized features. The algorithm is tested on a real world example, after which the results are discussed.