Abstract
On the basis of applying neutral networks to the track recognition problem the investigations are made according to the specific properties of such discrete detectors as multiwire proportional chambers. These investigations result in the modification of the so-called rotor model in a neutral neural network. The energy function of a network in this modification contains only one cost term. This speeds up calculations considerably. The reduction of the energy function is done by the neuron selection with the help of simplegeometrical and energetical criteria. Besides, the cellular automata were applied to preliminary selection of data that made it possible to create an initial network configuration with the energy closer to its global minimum. The algorithm was tested on 10{sup 4} real three-prong events obtained from the ARES-spectrometer. The results are satisfactory including the noise robustness and good resolution of nearby going tracks. 12 refs.; 10 figs.
Citation Formats
Glazov, A A, Kisel`, I V, Konotopskaya, E V, Neskoromnyj, V N, and Ososkov, G A.
Track reconstruction in discrete detectors by neutral networks.
JINR: N. p.,
1992.
Web.
Glazov, A A, Kisel`, I V, Konotopskaya, E V, Neskoromnyj, V N, & Ososkov, G A.
Track reconstruction in discrete detectors by neutral networks.
JINR.
Glazov, A A, Kisel`, I V, Konotopskaya, E V, Neskoromnyj, V N, and Ososkov, G A.
1992.
"Track reconstruction in discrete detectors by neutral networks."
JINR.
@misc{etde_10194158,
title = {Track reconstruction in discrete detectors by neutral networks}
author = {Glazov, A A, Kisel`, I V, Konotopskaya, E V, Neskoromnyj, V N, and Ososkov, G A}
abstractNote = {On the basis of applying neutral networks to the track recognition problem the investigations are made according to the specific properties of such discrete detectors as multiwire proportional chambers. These investigations result in the modification of the so-called rotor model in a neutral neural network. The energy function of a network in this modification contains only one cost term. This speeds up calculations considerably. The reduction of the energy function is done by the neuron selection with the help of simplegeometrical and energetical criteria. Besides, the cellular automata were applied to preliminary selection of data that made it possible to create an initial network configuration with the energy closer to its global minimum. The algorithm was tested on 10{sup 4} real three-prong events obtained from the ARES-spectrometer. The results are satisfactory including the noise robustness and good resolution of nearby going tracks. 12 refs.; 10 figs.}
place = {JINR}
year = {1992}
month = {Dec}
}
title = {Track reconstruction in discrete detectors by neutral networks}
author = {Glazov, A A, Kisel`, I V, Konotopskaya, E V, Neskoromnyj, V N, and Ososkov, G A}
abstractNote = {On the basis of applying neutral networks to the track recognition problem the investigations are made according to the specific properties of such discrete detectors as multiwire proportional chambers. These investigations result in the modification of the so-called rotor model in a neutral neural network. The energy function of a network in this modification contains only one cost term. This speeds up calculations considerably. The reduction of the energy function is done by the neuron selection with the help of simplegeometrical and energetical criteria. Besides, the cellular automata were applied to preliminary selection of data that made it possible to create an initial network configuration with the energy closer to its global minimum. The algorithm was tested on 10{sup 4} real three-prong events obtained from the ARES-spectrometer. The results are satisfactory including the noise robustness and good resolution of nearby going tracks. 12 refs.; 10 figs.}
place = {JINR}
year = {1992}
month = {Dec}
}