Minimum variance gain nonuniformity estimation in infrared focal plane array sensors

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

Abstract

In this paper, a minimum variance estimator for the gain nonuniformity (NU) in infrared (IR) focal plane array (FPA) imaging system is presented. Recently, we have developed a recursive filter estimator for the offset NU using only the collected scene data, assuming that the offset is a constant in a block of frames where it is estimated. The principal assumption of this scene-based NU correction (NUC) method is that the gain NU is a known constant and does not vary in time. However, in several FPA real systems the gain NU drift is significant. For this reason, in this work we present a gain NU drift estimation based on the offset NU recursive estimation assuming that gain and offset are jointly distributed. The efficacy of this NUC technique is demonstrated by employing several real infrared video se quences.

Original languageEnglish
Title of host publicationProgress in Pattern Recognition, Image Analysis, Computer Vision and Applications - 14th Iberoamerican Conference on Pattern Recognition, CIARP 2009, Proceedings
Pages1047-1053
Number of pages7
DOIs
StatePublished - 1 Dec 2009
Event14th Iberoamerican Conference on Pattern Recognition, CIARP 2009 - Guadalajara, Jalisco, Mexico
Duration: 15 Nov 200918 Nov 2009

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5856 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th Iberoamerican Conference on Pattern Recognition, CIARP 2009
CountryMexico
CityGuadalajara, Jalisco
Period15/11/0918/11/09

Keywords

  • Image sequence processing
  • Infrared focal plane arrays
  • Minimum variance estimator
  • Signal processing

Fingerprint Dive into the research topics of 'Minimum variance gain nonuniformity estimation in infrared focal plane array sensors'. Together they form a unique fingerprint.

Cite this