Advertisement

Linear Filters: Heuristic Theory and Stability

  • François Goudail
  • Philippe Réfrégier

Abstract

Linear filters are among the most widely used tools in image processing for such applications as target detection, localization and classification (or recognition). They are based on correlations between the searched objects (or a linear combination of them) and the analyzed scene. The correlation operation has several advantages in the context of image processing. First, it is an intrinsically position invariant recognition method which makes it possible to recognize an object whatever its location in the image. Second, correlation is quite robust to noise, and thus often constitutes an efficient way of processing very noisy images. Finally, it can be computed at a relatively low computational cost if fast methods are available, based for example on Fast Fourier Transform (FFT).

Keywords

Power Spectral Density Input Image Noisy Image Matched Filter Correlation Peak 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media New York 2004

Authors and Affiliations

  • François Goudail
    • 1
  • Philippe Réfrégier
    • 1
  1. 1.Fresnel InstituteENSPMMarseilleFrance

Personalised recommendations