CORNER ENHANCEMENT IN CURVATURE SCALE-SPACE

Corner detection is very important for pattern recognition, computer v ision and other works. However, successfully extracting corners from a planar curve is difficult because we do not know how to select the ap propriate scale for the extraction. In this paper, the problem of corn er enhancement via the curve scale space is addressed. The main idea o f the approach is to deform the original curve (or construct the curve scale space) and the corners on the deformed curve become distinguish ed from other structures which eases the process of corner detection. Started from the general geometric heat flow (GGHF), we study under wh at conditions the GGHF satisfies the scale space causality criteria. T his is very important because many different evolution approaches, whi ch have regular scale space properties and other specific properties, can then be found from these conditions. The criteria of corner enhanc ement are also proposed. Having all these constraints, a new curve evo lution scheme which can enhance strong corners suppress noise and sati sfies the scale space criteria is presented. (C) 1998 Pattern Recognit ion Society. Published by Elsevier Science Ltd. All rights reserved.