Real-Time Inference With 2D Convolutional Neural Networks on Field Programmable Gate Arrays for High-Rate Particle Imaging Detectors.
data acquisition system
data selection
fast machine vision
hardware acceleration of deep learning
liquid argon time projection chamber
particle imaging
real-time machine leaning
trigger system
Journal
Frontiers in artificial intelligence
ISSN: 2624-8212
Titre abrégé: Front Artif Intell
Pays: Switzerland
ID NLM: 101770551
Informations de publication
Date de publication:
2022
2022
Historique:
received:
14
01
2022
accepted:
12
04
2022
entrez:
6
6
2022
pubmed:
7
6
2022
medline:
7
6
2022
Statut:
epublish
Résumé
We present a custom implementation of a 2D Convolutional Neural Network (CNN) as a viable application for real-time data selection in high-resolution and high-rate particle imaging detectors, making use of hardware acceleration in high-end Field Programmable Gate Arrays (FPGAs). To meet FPGA resource constraints, a two-layer CNN is optimized for accuracy and latency with KerasTuner, and network
Identifiants
pubmed: 35664508
doi: 10.3389/frai.2022.855184
pmc: PMC9157595
doi:
Types de publication
Journal Article
Langues
eng
Pagination
855184Informations de copyright
Copyright © 2022 Jwa, Di Guglielmo, Arnold, Carloni and Karagiorgi.
Déclaration de conflit d'intérêts
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Références
Front Big Data. 2022 Mar 23;5:828666
pubmed: 35402906
Front Artif Intell. 2021 Aug 24;4:649917
pubmed: 34505055
Front Big Data. 2021 Jan 12;3:598927
pubmed: 33791596
Front Artif Intell. 2021 Jul 09;4:676564
pubmed: 34308339
Eur Phys J C Part Fields. 2021;81(4):322
pubmed: 34720713
Nature. 2018 Aug;560(7716):41-48
pubmed: 30068955