Title :
An architecture for feature detection utilizing dynamic synapses
Author :
Heittmann, Arne ; Ramacher, Ulrich
Author_Institution :
Corporate Res. Syst. Technol., Infineon Technol. AG, Munich, Germany
Abstract :
An architecture for feature detection based on artificial pulsed neural networks is presented. By utilizing substancial principles like correlation of pulses using dynamic synapses as well as distribution of connections defined by connection densities, arbitrary detectors up to a complexity of simple cells can be implemented.
Keywords :
computer architecture; feature extraction; neural nets; arbitrary detectors; artificial pulsed neural networks; connection density; dynamic synapses; feature detection; pulse correlation; Biological information theory; Biological systems; Computer vision; Detectors; Equations; Information processing; Neural networks; Neurons; Robustness; Surveillance;
Conference_Titel :
Circuits and Systems, 2004. MWSCAS '04. The 2004 47th Midwest Symposium on
Print_ISBN :
0-7803-8346-X
DOI :
10.1109/MWSCAS.2004.1354171
