Machine learning on compton event identification for a nano-satellite mission

  • Haitao CaoEmail author
  • Denis Bastieri
  • Riccardo Rando
  • Giorgio Urso
  • Gaoyong Luo
  • Alessandro Paccagnella
Original Article


Nano-satellite MeV telescope is becoming attractive nowadays. The dominant interaction mechanism of the electromagnetic spectrum around 1MeV is Compton scattering. However, the gamma-rays generated by primary particles hitting the atmosphere and the pair production events are the two significant background events when the satellite is operating in Low Earth Orbit. In this paper, we applied Machine Learning models to identify and reject the two troublesome background event types. Ensemble technique and imbalance solution are explored in order to obtain a better performance. Experiments demonstrated that the proposed methods can discriminate the pair events with a high accuracy, and the satellite’s sensitivity has also been improved dramatically.


Machine learning Neural network Ensemble methods Imbalance problem MeV telescope Loss functions 



Author Haitao Cao would like to acknowledge the scholarship supported by Guangzhou University, China and the excellent research facilities provided by Istituto Nazionale di Fisica Nucleare, Padova, Italy.


  1. 1.
    Abadi, M., Barham, P., Chen, J., et al.: TensorFlow: A system for large-scale machine learning. 12th USENIX Symp. Operating Syst. Des. Implement 16, 265–283 (2016)Google Scholar
  2. 2.
    Andritschke, R., Zoglauer, A., Kanbach, G., et al.: The Compton and pair creation telescope MEGA. Exp. Astron. 20(1-3), 395–403 (2005)ADSCrossRefGoogle Scholar
  3. 3.
    Atwood, W., Abdo, A.A., Ackermann, M., et al.: The large area telescope on the Fermi gamma-ray space telescope mission. Astrophys. J. 697(2), 1071 (2009)ADSCrossRefGoogle Scholar
  4. 4.
    Berlato, F.: Design and Optimization Around 1 MeV of a Calorimeter for a Cubesat Mission. Università degli Studi di Padova, Master’s thesis (2016)Google Scholar
  5. 5.
    De Angelis, A., Tatischeff, V., Tavani, M., et al.: The e-ASTROGAM mission. Exp. Astron. 44(1), 25–82 (2017)ADSCrossRefGoogle Scholar
  6. 6.
    Denton, E.L., Chintala, S., Fergus, R., et al.: Deep generative image models using a Laplacian pyramid of adversarial networks. Adv. Neural Inf. Process. Syst, 1486–1494 (2015)Google Scholar
  7. 7.
    Dietterich, T.G.: Ensemble learning. Handbook Brain Theory Neural Netw. 2, 110–125 (2002)Google Scholar
  8. 8.
    Freund, Y., Schapire, R.E.: A desicion-theoretic generalization of on-line learning and an application to boosting. In: European Conference on Computational Learning Theory, pp. 23–37 (1995)Google Scholar
  9. 9.
    Garrido Beltrán, L., Gascón Fora, D., Graciani Díaz, R., et al.: Design concepts for the Cherenkov telescope array CTA: An advanced facility for ground-based high-energy gamma-ray astronomy. Exp. Astron. 32(3), 193–316 (2011)ADSCrossRefGoogle Scholar
  10. 10.
    Gehrels, N., Chincarini, G., Giommi, P., et al.: The Swift gamma-ray burst mission. Astrophys. J. 611(2), 1005 (2004)ADSCrossRefGoogle Scholar
  11. 11.
    Géron, A.: Hands-on machine learning with Scikit-Learn and TensorFlow: Concepts, tools and techniques to build intelligent systems (2017)Google Scholar
  12. 12.
    Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the 13th International Conference on Artificial Intelligence and Statistics, pp. 249–256 (2010)Google Scholar
  13. 13.
    Goodfellow, I., Pouget-Abadie, J., Mirza, M., et al.: Generative adversarial nets. In: Advances in Neural Information Processing Systems, pp. 2672–2680 (2014)Google Scholar
  14. 14.
    Harrison, F.A., Craig, W.W., Christensen, F.E., et al.: The nuclear spectroscopic telescope array (NuSTAR) high-energy X-ray mission. Astrophys. J. 770(2), 103 (2013)ADSCrossRefGoogle Scholar
  15. 15.
    Hinton, J.: Ground-based gamma-ray astronomy with Cherenkov telescopes. J. Phys. 11(5), 055005 (2009)Google Scholar
  16. 16.
    Ioffe, S., Szegedy, C.: Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: International Conference on Machine Learning, pp. 448–456 (2015)Google Scholar
  17. 17.
    Janocha, K., Czarnecki, W.M.: On loss functions for deep neural networks in classification. Schedae Informaticae 25, 49–59 (2016)Google Scholar
  18. 18.
    Japkowicz, N., Stephen, S.: The class imbalance problem: A systematic study. Intell. Data Anal. 6(5), 429–449 (2002)CrossRefzbMATHGoogle Scholar
  19. 19.
    Jiménez, D.: Dynamically weighted ensemble neural networks for classification. In: IEEE International Joint Conference on Neural Networks Proceedings, vol. 1, pp. 753–756. IEEE (1998)Google Scholar
  20. 20.
    Kam, H.T.: Random decision forest. In: Proceedings of the 3rd International Conference on Document Analysis and Recognition, pp. 14–18 (1995)Google Scholar
  21. 21.
    Lucchetta, G.: Design and Optimization Around 1 MeV of a Tracker for a CubeSat Mission. Università degli Studi di Padova, Master’s thesis (2016)Google Scholar
  22. 22.
    Lucchetta, G., Berlato, F., Rando, R., et al.: Scientific performance of a nano-satellite MeV telescope. Astron. J. 153(5), 237 (2017)ADSCrossRefGoogle Scholar
  23. 23.
    Mizumoto, T., Matsuoka, Y., Mizumura, Y., et al.: New readout and data-acquisition system in an electron-tracking Compton camera for MeV gamma-ray astronomy (SMILE-II). Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 800, 40–50 (2015)ADSCrossRefGoogle Scholar
  24. 24.
    Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. In: Proceedings of the 27th International Conference on Machine Learning, pp. 807–814 (2010)Google Scholar
  25. 25.
    Odena, A., Olah, C., Shlens, J.: Conditional image synthesis with auxiliary classifier GANs. In: Proceedings of the 34th International Conference on Machine Learning, vol. 70, pp. 2642–2651 (2017)Google Scholar
  26. 26.
    Pedregosa, F., Varoquaux, G., Gramfort, A., et al.: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)MathSciNetzbMATHGoogle Scholar
  27. 27.
    Rumelhart, D.E., Hinton, G.E., Williams, R.J., et al.: Learning representations by back-propagating errors. Cogn. Model. 5(3), 1 (1988)zbMATHGoogle Scholar
  28. 28.
    Salimans, T., Goodfellow, I., Zaremba, W., et al.: Improved techniques for training GANs. Adv. Neural Inf. Process. Syst., 2234–2242 (2016)Google Scholar
  29. 29.
    Schoenfelder, V., Aarts, H., Bennett, K., et al.: Instrument description and performance of the imaging gamma-ray telescope COMPTEL aboard the Compton gamma-ray observatory. Astrophys. J. Suppl. Ser. 86, 657–692 (1993)ADSCrossRefGoogle Scholar
  30. 30.
    Sønderby, C.K., Caballero, J., Theis, L., et al.: Amortised map inference for image super-resolution. International Conference on Learning Representations (2017)Google Scholar
  31. 31.
    Srivastava, N., Hinton, G.E., Krizhevsky, A., et al.: Dropout: A simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  32. 32.
    Sutskever, I., Martens, J., Dahl, G., et al.: On the importance of initialization and momentum in deep learning. In: International Conference on Machine Learning, pp. 1139–1147 (2013)Google Scholar
  33. 33.
    Tavani, M., Barbiellini, G., Argan, A., et al.: The AGILE mission. Astron. Astrophys. 502(3), 995–1013 (2009)ADSCrossRefGoogle Scholar
  34. 34.
    Winkler, C., Diehl, R., Ubertini, P., et al.: INTEGRAL: Science highlights and future prospects. Space Sci. Rev. 161(1–4), 149–177 (2011)ADSCrossRefGoogle Scholar
  35. 35.
    Zhang, Y., Zhang, H., Cai, J., et al.: A weighted voting classifier based on differential evolution. Abstract Appl. Anal., 2014 (2014)Google Scholar
  36. 36.
    Zoglauer, A., Andritschke, R., Boggs, S.E., et al.: MEGAlib–simulation and data analysis for low-to-medium-energy gamma-ray telescopes. In: Proceedings of the International Society for Optical Engineering, vol. 7011, 70113F (2008)Google Scholar
  37. 37.
    Zoglauer, A.C.: First Light for the Next Generation of Compton and Pair Telescopes. Ph.D. thesis, Technische Universität München (2005)Google Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.Department of Information EngineeringUniversity of PadovaPadovaItaly
  2. 2.Istituto Nazionale di Fisica NuclearePadovaItaly
  3. 3.School of Physics and Electronic EngineeringGuangzhou UniversityGuangzhouChina
  4. 4.Department of Physics and Astronomy “G. Galilei”University of PadovaPadovaItaly
  5. 5.Center for AstrophysicsGuangzhou UniversityGuangzhouChina

Personalised recommendations