To main content

A modular hyperspectral image processing pipeline for cubesats

Abstract

This work describes the onboard image processing pipeline that is designed to be used onboard a CubeSat with a Hyper Spectral (Imager/Image) (HSI), namely the HYPSO-I satellite. While the system itself supports both the use of a Field Programmable Gate Array (FPGA) and CPUs for processing, this work focuses on the latter. A reference to the source code is also enclosed within the text. Some of the planned and developed modules are presented, along with the design choices needed to accommodate the computationally constrained system. The execution time of the pipeline on target hardware with representative hyperspectral data is also shown. The results indicate that the modular pipeline framework is suitable for deployment onboard HYPSO-I.
Read the publication

Category

Academic article

Language

English

Author(s)

  • Sivert Bakken
  • Aksel Skrøvset Danielsen
  • Kristine Døsvik
  • Joseph Garrett
  • Milica Orlandic
  • Dennis D. Langer
  • Tor Arne Johansen

Affiliation

  • SINTEF Ocean / Fisheries and New Biomarine Industry
  • Norwegian University of Science and Technology

Year

2022

Published in

Workshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing

ISSN

2158-6276

View this publication at Norwegian Research Information Repository