Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Comparison between orthogonal subspace projection and background subtraction techniques applied to remote-sensing

Avishai Ben-David1, Hsuan Ren

  • 1Edgewood Chemical Biological Center, Aberdeen Proving Ground, Maryland 21010, USA. avishai.bendavid@us.army.mil

Applied Optics
|July 2, 2005
PubMed
Summary

Two methods for solving spectral abundance (alpha) were compared: subtraction and orthogonal subspace projection (OSP). The noise angle between background (B) and noise (n) determines which spectral analysis method, subtraction or OSP, offers superior results.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Hyperspectral Tensor Completion Using Low-Rank Modeling and Convex Functional Analysis.

IEEE transactions on neural networks and learning systems·2023
Same author

Estimation method for serial dilution experiments.

Journal of microbiological methods·2014
Same author

Probability theory for 3-layer remote sensing in ideal gas law environment.

Optics express·2013
Same author

Regression model for estimating inactivation of microbial aerosols by solar radiation.

Photochemistry and photobiology·2013
Same author

Probability theory for 3-layer remote sensing radiative transfer model: univariate case.

Optics express·2012
Same author

A model for inactivation of microbes suspended in the atmosphere by solar ultraviolet radiation.

Photochemistry and photobiology·2010

Area of Science:

  • Spectral analysis
  • Hyperspectral imaging
  • Signal processing

Background:

  • Spectral measurements (r) are often modeled as a sum of a spectral target vector (d) weighted by its abundance (alpha), a stochastic spectral background vector (B), and noise (n).
  • Accurate determination of alpha is crucial in various applications, including remote sensing and chemical analysis.

Purpose of the Study:

  • To compare the performance of two methods for solving for spectral abundance (alpha): subtraction and orthogonal subspace projection (OSP).
  • To investigate the influence of the geometric relationship between the spectral background (B) and noise (n) on the efficacy of each solution.

Main Methods:

  • Solving the measurement equation r = B + alphad + n for alpha using two distinct approaches: direct background subtraction and orthogonal subspace projection (OSP).

Related Experiment Videos

  • Analyzing the geometric properties of the solutions, particularly the noise vector's orientation relative to the background vector.
  • Utilizing simulations and one-dimensional hyperspectral measurements to validate the findings.

    • Main Results:

    • The superiority of a solution depends on the angular distribution of the noise relative to the background.
    • When the noise-angle distribution is uniform, the subtraction solution consistently outperforms OSP.
    • The OSP solution becomes advantageous when noise is concentrated in directions orthogonal to the background vector.

    Conclusions:

    • The choice between subtraction and OSP for spectral abundance estimation is critically dependent on the noise characteristics.
    • Understanding the noise-angle distribution is key to predicting and selecting the optimal spectral analysis method.
    • The study provides practical insights for improving spectral data analysis in the presence of background and noise.