Chapter 13: Automatic scale selection

Chapter 13 in Scale-Space Theory in Computer Vision proposes a general heuristic principle for scale selection stating that local extrema over scales of different combinations of normalized scale invariant derivatives are likely candidates to correspond to interesting structures. The resulting methodology lends itself naturally to two-stage algorithms; feature detection at coarse scales followed by feature localization at finer scales. Support is given by theoretical considerations and experiments on blob detection, junction detection, and edge detection.
Responsible for this page: Tony Lindeberg