Community Research and Development Information Service - CORDIS

Abstract

A hybrid segmentation method has been developed which integrates edge detection and region growing in order to overcome their respective weaknesses. The method consists of the following stages: (i) filtering; (ii) edge detection and following; (iii) edge fragment linking; (iv) region growing. In (ii) edge detection is carried out. The resulting edge magnitude values are thresholded and on the thresholded values a thinning operation is performed in order to create one pixel thick edges. In (iii) the resulting edge fragments are linked together where possible by detecting one pixel wide gaps between edge fragments. By connecting the edge fragments, closed polygons are formed, dividing the image into a set of sub-images. Edge fragments not belonging to a closed polygon are pruned. In (iv) region growing is carried out within every polygon. Regions are not allowed to grow outside the polygons. The region growing method used is the iterative best merge approach, which merges in each scan over the image the pixel/region pair with a lowest cost value. For merging remaining isolated pixels context rules are employed. Comparative results are shown of a non-segmented Landsat-TM scene and its segmented counterpart classified by an artificial neural network. Moreover the use of the segmentation for filtering SAR imagery is indicated.

Additional information

Authors: SCHOENMAKERS R P H M, JRC Ispra (IT);WILKINSON G G, JRC Ispra (IT);SCHOUTEN T E, Katholieke Universiteit Nijmegen (NL)
Bibliographic Reference: Paper presented: SPIE, Rome (IT), September 26-30 1994
Availability: Available from (1) as Paper EN 38573 ORA
Record Number: 199511532 / Last updated on: 1995-12-12
Category: PUBLICATION
Original language: en
Available languages: en
Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top