Szczegóły publikacji
Opis bibliograficzny
Hybrid pattern recognition for charged particle tracking: Hough transform and convolutional neural networks / Carlo VARNI, Krzysztof CIEŚLA, Marcin Wolter, Radoslaw Karol Mieńkowski, Noemi Calace, Tomasz BOŁD // Nuclear Instruments & Methods in Physics Research . Section A, Accelerators, spectrometers, detectors and associated equipment ; ISSN 0168-9002. — 2027 — vol. 1094 art. no. 172140, s. 1-11. — Bibliogr. s. 11, Abstr. — Publikacja dostępna online od: 2026-10-11
Autorzy (6)
- AGHVarni Carlo
- AGHCieśla Krzysztof
- Wolter Marcin
- AGHMieńkowski Radosław
- Calace Noemi
- AGHBołd Tomasz
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 171130 |
|---|---|
| Data dodania do BaDAP | 2026-11-06 |
| Tekst źródłowy | URL |
| DOI | 10.1016/j.nima.2026.172140 |
| Rok publikacji | 2027 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp |  |
| Creative Commons | |
| Czasopismo/seria | Nuclear Instruments & Methods in Physics Research, Section A, Accelerators Spectrometers, Detectors and Associated Equipment |
Abstract
Reconstructing charged-particle tracks in silicon detectors is a central task in high-energy physics experiments and a key component of both offline reconstruction and online event selection. Within the reconstruction chain, the efficient and high-purity formation of track candidates plays a critical role in the overall performance. Among the many approaches developed over the years, the Hough transform (HT) has been widely studied as a fast geometry-driven method for track finding. However, in high-occupancy environments such as those expected at the High-Luminosity LHC (HL-LHC), the HT tends to produce a large number of spurious candidates, leading to increased computational overhead in subsequent reconstruction stages. In this work, we present a hybrid approach in which the HT serves as a first-stage data preparation step, providing an image of its parameter space as an input to a neural network trained to suppress false track candidates. The method combines the speed of the HT with the discriminative power of machine learning to achieve both efficiency and purity. Moreover, no data transformations are required to combine the two stages, resulting in a simpler and potentially more performant algorithm. Performance studies using the Open Data Detector simulated in the ACTS framework under realistic HL-LHC pile-up conditions will be presented.
