Reduced resolution and scale space for dominant feature detection in contours

Analysis of contours or curves at various scales or levels of smoothing (sc ale space) is an important tool in curve segmentation and feature detection . In this paper we propose the application of the hierarchical discrete cor relation algorithm for efficiently calculating and creating a scale space o f curves. In conjunction with this procedure we also investigate the use of the reduced resolution or Gaussian pyramid representation of the set of sm oothed curves as the basis for initially localizing and detecting features. We also create an approximation to the above algorithms that is computatio nally less expensive. Finally, we propose a new inter-scale method for curv e segmentation and feature detection based on the motion of a curve through scale space. (C) 2000 Pattern Recognition Society. Published by Elsevier S cience Ltd. All rights reserved.