Detecting Salient Blob-Like Image Structures and Their Scales with a Scale-Space Primal Sketch: A Method for Focus-of-Attention

Tony Lindeberg

Technical report ISRN KTH/NA/P--93/33--SE.

Also in International Journal of Computer Vision, 11(3), 283--318, 1993.

Abstract

This article presents:

a multi-scale representation of grey-level shape called the scale-space primal sketch, which makes explicit both features in scale-space and the relations between structures at different scales,
a methodology for extracting significant blob-like image structures from this representations, and
applications to edge detection, histogram analysis, and junction classification demonstrating how the proposed method can be used for guiding later stage visual processes.

The representation gives a qualitative description of image structure, which allows for detection of stable scales and associated regions of interest in a solely bottom-up data-driven way. In other words, it generates coarse segmentation cues, and can hence be seen as preceding further processing, which can then be properly tuned. It is argued that once such information is available, many other processing tasks can become much simpler. Experiments on real imagery demonstrate that the proposed theory gives intuitive results.

Full paper: (PostScript 1.7Mb) (PDF 1.7Mb)

The underlying algorithm for linking blobs (including local extrema and critical points) across scales and registering bifurcation events (singularity points) is described in chapter 9 in Lindeberg (1994) "Scale-Space Theory for Computer Vision", Kluwer Academic Publishers.

Background and related material: (Monograph on scale-space theory) (Theory for relating image structures over scales "deep structure") (Alternative method for multi-scale blob detection and more general methodology for feature detection with automatic scale selection) (Other publications on scale-space theory) (Encyclopedia entry on scale-space theory)

Tony Lindeberg <tony@bion.kth.se>