Computational methods for improving thermal imaging for consumer devices

Research output: Chapter in Book or Conference Publication/ProceedingConference Publicationpeer-review

11 Citations (Scopus)

Abstract

In consumer imaging, the spatial resolution of thermal microbolometer arrays is limited by the large physical size of the individual detector elements. This also limits the number of pixels per image. If thermal sensors are to find a place in consumer imaging, as the newly released FLIR One would suggest, this resolution issue must be addressed. Our work focuses on improving the output quality of low resolution thermal cameras through computational means. The method we propose utilises sub-pixel shifts and temporal variations in the scene, using information from thermal and visible channels. Results from simulations and lab experiments are presented.

Original languageEnglish
Title of host publicationThermosense
Subtitle of host publicationThermal Infrared Applications XXXVII
EditorsSheng-Jen Hsieh, Joseph N. Zalameda
PublisherSPIE
ISBN (Electronic)9781628416015
DOIs
Publication statusPublished - 2015
EventSPIE ThermoSense Conference: Thermal Infrared Applications XXXVII - Baltimore, United States
Duration: 20 Apr 201523 Apr 2015

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume9485
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceSPIE ThermoSense Conference: Thermal Infrared Applications XXXVII
Country/TerritoryUnited States
CityBaltimore
Period20/04/1523/04/15

Keywords

  • Computational imaging
  • Diffraction limit
  • Optics
  • Spatial frequencies
  • Super-resolution
  • Temporal
  • Thermal imaging

Fingerprint

Dive into the research topics of 'Computational methods for improving thermal imaging for consumer devices'. Together they form a unique fingerprint.

Cite this