基于神经网络识别 Mini-EUSO 望远镜数据中的流星迹
天体物理仪器与方法
2023-11-28 v1 计算机视觉与模式识别
机器学习
摘要
Mini-EUSO 是一台宽角荧光望远镜，从国际空间站记录地球夜间大气中的紫外（UV）辐射。流星是多种现象之一，不仅在可见光范围也在 UV 中显现。我们提出两个简单的人工神经网络，就二分类问题以高精度识别 Mini-EUSO 数据中的流星信号。我们预期类似架构可有效用于其他荧光望远镜的信号识别，无论信号性质如何。因其简单性，这些网络可部署于未来轨道或气球实验的星载电子设备中。
引用
@article{arxiv.2311.14983,
title = {Neural Network Based Approach to Recognition of Meteor Tracks in the Mini-EUSO Telescope Data},
author = {Mikhail Zotov and Dmitry Anzhiganov and Aleksandr Kryazhenkov and Dario Barghini and Matteo Battisti and Alexander Belov and Mario Bertaina and Marta Bianciotto and Francesca Bisconti and Carl Blaksley and Sylvie Blin and Giorgio Cambiè and Francesca Capel and Marco Casolino and Toshikazu Ebisuzaki and Johannes Eser and Francesco Fenu and Massimo Alberto Franceschi and Alessio Golzio and Philippe Gorodetzky and Fumiyoshi Kajino and Hiroshi Kasuga and Pavel Klimov and Massimiliano Manfrin and Laura Marcelli and Hiroko Miyamoto and Alexey Murashov and Tommaso Napolitano and Hiroshi Ohmori and Angela Olinto and Etienne Parizot and Piergiorgio Picozza and Lech Wiktor Piotrowski and Zbigniew Plebaniak and Guillaume Prévôt and Enzo Reali and Marco Ricci and Giulia Romoli and Naoto Sakaki and Kenji Shinozaki and Christophe De La Taille and Yoshiyuki Takizawa and Michal Vrábel and Lawrence Wiencke},
journal= {arXiv preprint arXiv:2311.14983},
year = {2023}
}
备注
15 pages