Algorithm for computing the scale-space primal sketch

  • Tony Lindeberg
Part of the The Springer International Series in Engineering and Computer Science book series (SECS, volume 256)


When building a representation like the scale-space primal sketch, several computational aspects must be considered. Algorithms are needed for smoothing grey-level images, detecting grey-level blobs in smoothed grey-level images, registering bifurcations, linking grey-level blobs across scales into scale-space blobs, and computing the scale-space blob volumes. In this chapter how this can be done will be briefly described. Before starting, it should be remarked that some algorithmic descriptions will, by necessity, be somewhat technical. Therefore, the reader is recommended to keep the overview description from chapter 7 in mind when penetrating this material. It should also be emphasized that, although the treatment deals mainly with the two-dimensional case, the ideas behind it are general and can (with appropriate modifications) be applied both in lower and higher dimensions.


Scale Level Local Extremum Coarse Scale Coarse Level Support Region 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer Science+Business Media Dordrecht 1994

Authors and Affiliations

  • Tony Lindeberg
    • 1
  1. 1.Royal Institute of TechnologyStockholmSweden

Personalised recommendations