An Orientation Selective Neural Network for Pattern Identification in Particle Detectors
Abstract
We present an algorithm for identifying linear patterns on a two(cid:173) dimensional lattice based on the concept of an orientation selective cell, a concept borrowed from neurobiology of vision. Construct(cid:173) ing a multi-layered neural network with fixed architecture which implements orientation selectivity, we define output elements cor(cid:173) responding to different orientations, which allow us to make a se(cid:173) lection decision. The algorithm takes into account the granularity of the lattice as well as the presence of noise and inefficiencies. The method is applied to a sample of data collected with the ZEUS detector at HERA in order to identify cosmic muons that leave a linear pattern of signals in the segmented calorimeter. A two dimensional representation of the relevant part of the detector is used. The algorithm performs very well. Given its architecture, this system becomes a good candidate for fast pattern recognition in parallel processing devices.
Cite
Text
Abramowicz et al. "An Orientation Selective Neural Network for Pattern Identification in Particle Detectors." Neural Information Processing Systems, 1996.Markdown
[Abramowicz et al. "An Orientation Selective Neural Network for Pattern Identification in Particle Detectors." Neural Information Processing Systems, 1996.](https://mlanthology.org/neurips/1996/abramowicz1996neurips-orientation/)BibTeX
@inproceedings{abramowicz1996neurips-orientation,
title = {{An Orientation Selective Neural Network for Pattern Identification in Particle Detectors}},
author = {Abramowicz, Halina and Horn, David and Naftaly, Ury and Sahar-Pikielny, Carmit},
booktitle = {Neural Information Processing Systems},
year = {1996},
pages = {925-931},
url = {https://mlanthology.org/neurips/1996/abramowicz1996neurips-orientation/}
}