On-the-fly data pipeline for image processing enabling real-time persistence correction

Christopher Mandla, Muhammad Subhan Hameed, Viseslav Aracic, Sabine Ott, Markus Plattner, Andreas Herkersdorf, Thomas Wild

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, we present our concept of an on-the-fly image processing pipeline for ground-based telescopes. Our focus lies on the correction of undesired but unavoidable detector effects like persistence. State of the art systems like ESO's NGC perform operation of detectors and acquisition of science data. Instrument operators monitor the science data stream during measurement. Detector issues like persistence worsen the image quality and impair the inspection tasks. To overcome these difficulties, we develop the real-time data processing system EIDAS (Enhanced Infrared Detector Acquisition System) as an add-on to the NGC. Therewith, detector effects in the raw image stream are corrected on-the-fly and astronomical objects are visualized more clearly for direct inspections. EIDAS accompanies CPU, FPGA and GPU as well as algorithms optimized for this hardware architecture.

Original languageEnglish
Title of host publicationX-Ray, Optical, and Infrared Detectors for Astronomy IX
EditorsAndrew D. Holland, James Beletic
PublisherSPIE
ISBN (Electronic)9781510636958
DOIs
StatePublished - 2020
EventX-Ray, Optical, and Infrared Detectors for Astronomy IX 2020 - Virtual, Online, United States
Duration: 14 Dec 202022 Dec 2020

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11454
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceX-Ray, Optical, and Infrared Detectors for Astronomy IX 2020
Country/TerritoryUnited States
CityVirtual, Online
Period14/12/2022/12/20

Keywords

  • Data Pipeline
  • Data Processing
  • FPGA
  • GPU
  • Infrared Detectors
  • Real-time

Fingerprint

Dive into the research topics of 'On-the-fly data pipeline for image processing enabling real-time persistence correction'. Together they form a unique fingerprint.

Cite this