Radiometric correction of hyperspectral imaging data in spacial dimension and spectral dimension

  • Hongying Liu*
  • , Yana Guan
  • , Qingli Li
  • , Jingao Liu
  • , Yongqi Xue
  • *Corresponding author for this work

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

5 Scopus citations

Abstract

Because sample are affected by light, lens and sensor in optical channel in hyperspectral imaging system, image quality and spectrums are also affected. The paper proposes the gray correction coefficient algorithm with spatial dimension and spectral dimension to preprocess molecular hyperspectral imaging data. The experimental results show that the algorithm can carry out radiometric correction in spatial dimension and spectral dimension and eliminate the effects of light, lens and sensor. Image quality is significantly improved. The spectrum curves indicating true biochemical characteristics of the sample are extracted that classification results based on spectral information are better than uncorrected sample.

Original languageEnglish
Title of host publication2011 2nd International Conference on Mechanic Automation and Control Engineering, MACE 2011 - Proceedings
Pages4265-4268
Number of pages4
DOIs
StatePublished - 2011
Event2011 2nd International Conference on Mechanic Automation and Control Engineering, MACE 2011 - Inner Mongolia, China
Duration: 15 Jul 201117 Jul 2011

Publication series

Name2011 2nd International Conference on Mechanic Automation and Control Engineering, MACE 2011 - Proceedings

Conference

Conference2011 2nd International Conference on Mechanic Automation and Control Engineering, MACE 2011
Country/TerritoryChina
CityInner Mongolia
Period15/07/1117/07/11

Keywords

  • Molecular hyperspctral imaging(MHSI)
  • Radiometric correction
  • Spectral angle mapper

Fingerprint

Dive into the research topics of 'Radiometric correction of hyperspectral imaging data in spacial dimension and spectral dimension'. Together they form a unique fingerprint.

Cite this