Using GPUs to Speed up a Tomographic Reconstructor Based on Machine Learning

  1. Carlos González-Gutiérrez 1
  2. Jesús Daniel Santos-Rodríguez 1
  3. Ramón Ángel Fernández Díaz 2
  4. Jose Luis Calvo Rolle 3
  5. Nieves Roqueñí Gutiérrez 1
  6. Cos Juez, Francisco Javier de 1
  1. 1 Universidad de Oviedo

    Universidad de Oviedo

    Oviedo, España


  2. 2 Universidad de León

    Universidad de León

    León, España


  3. 3 Universidade da Coruña

    Universidade da Coruña

    La Coruña, España


International Joint Conference SOCO’16-CISIS’16-ICEUTE’16: San Sebastián, Spain, October 19th-21st, 2016 Proceedings
  1. Manuel Graña (coord.)
  2. José Manuel López-Guede (coord.)
  3. Oier Etxaniz (coord.)
  4. Álvaro Herrero (coord.)
  5. Héctor Quintián (coord.)
  6. Emilio Corchado (coord.)

Publisher: Springer Suiza

ISBN: 978-3-319-47364-2 3-319-47364-6 978-3-319-47363-5 3-319-47363-8

Year of publication: 2017

Pages: 279-289

Congress: International Conference on Computational Intelligence in Security for Information Systems (9. 2016. San Sebastián)

Type: Conference paper


The next generation of adaptive optics (AO) systems require tomographic techniques in order to correct for atmospheric turbulence along lines of sight separated from the guide stars. Multi-object adaptive optics(MOAO) is one such technique. Here we present an improved version of CARMEN, a tomographic reconstructor based on machine learning, using a dedicated neural network framework as Torch. We can observe a significant improvement on the training an execution times of the neural network, thanks to the use of the GPU.