Edge location with electrostatic region attractors

Author

Wilson, Richard C. ; Cross, Andrew D J ; Hancock, Edwin R.

Author_Institution

Dept. of Comput. Sci., York Univ., UK

Volume

2

fYear

1998

fDate

4-7 Oct 1998

Firstpage

535

Abstract

This paper describes physically based model for detecting closed edge-contours. According to our model, the magnitude of the Canny edge-gradient represents a raw charge-density. Edge segmentation is realised by considering the damped-motion of charged test-particles in the resulting electric field. The attractor regions of the electric field represent continuous regions that are separated by closed edge-contours. In order to strike a compromise between over and under smoothing the resulting edge-map, we provide a hierarchical algorithm for controlling the dynamics of the region attractors. We illustrate the method on both synthetic and real world data

Keywords

edge detection; electrostatics; image segmentation; Canny edge-gradient magnitude; attractor regions; charge-density; charged test-particles; closed edge-contours detection; continuous regions; damped-motion; edge location; edge segmentation; edge-map; electric field; electrostatic region attractors; hierarchical algorithm; over smoothing; physically based model; real world data; synthetic data; under smoothing; Computational modeling; Computer science; Detectors; Electrostatics; Hysteresis; Image edge detection; Image sampling; Joining processes; Smoothing methods; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on

Conference_Location

Chicago, IL

Print_ISBN

0-8186-8821-1

Type

conf

DOI

10.1109/ICIP.1998.723499

Filename

723499