An edge-based segmentation technique allowing to process very large range images quickly was developed. The proposed technique consists of two stages. First, a binary edge map is generated, then, a contour detection strategy is responsible for the extraction of the different boundaries. The first stage produces a binary edge map based on a scan line approximation technique. There is a difference with the previous techniques, as only two orthogonal scan line directions are considered. The planar curves defined by the elements contained in each scan line are approximated by oriented quadratic curves. The representative points from each curve are used to define a binary edge map. The second stage consists in a new approach to the classical contour extraction problem.
It shows a difference with the previous approaches which use the enclosed surface information; with the suggested technique, boundaries are obtained by using only the information contained into the binary edge map. It consists in linking the edge points by applying a graph strategy. Below, an illustration showing the boundaries obtained by using the proposed technique is given
A. D. Sappa and M. Devy, ``Efficient Contour Extraction in Range Image Segmentation for Building Modelling", Int. Symp. on Virtual and Augmented Architecture, Dublin, Ireland, June 2001.
A. D. Sappa, V. Bevilacqua and M. Devy, ``Improving a Genetic Algorithm Segmentation by means of a Fast Edge Detection Technique", IEEE Int. Conf. on Image Processing, Thessaloniki, Greece, October 2001 (submitted paper).
A. D. Sappa and M. Devy, ``Range Image Segmentation: Contour and Region Extraction by means of Minimum Spanning Tree Algorithm", IEEE Int. Conf. on Computer Vision, Vancouver, Canada, July 2001 (submitted paper).
A. D. Sappa and M. Devy, ``Segmentation d'images 3D par extraction des contours 3D", ORASIS 2001, Cahors, France, June 2001 (in French-accepted paper).
A. D. Sappa and M. Devy, ``Fast Range Image Segmentation by an Edge Detection Strategy", Third International Conference on 3D Digital Imaging and Modeling, Québec, Canada, June 2001 (accepted paper).
Here we see the contours extracted from a panoramic range image defined by 1400x8000 points.
