Thin nets extraction using a multi-scale approach

Thin nets are the lines where the grey level function is locally extremum i n a given direction. Recently, we have shown that it is possible to charact erize the thin nets using differential properties of the image surface. How ever, the method failed when these structures present different widths. In this paper we show that the extraction process of the thin nets, having dif ferent width, requires a multiscale analysis of the image. To design the fu sion process of the multiscale information, we will study the behavior of t he differential properties of the image surface, in particular the curvatur es, in scale space. We illustrate the efficiency of the proposed multi-scal e approach by extracting roads and blood vessels of different widths in sat ellite and medical images. (C) 1999 Academic Press.