Thin nets and Crest lines : Application to Satellite Data and Medical Images
Abstract
In this paper, we describe a new approach for extracting {\em thin nets} in grey level images. The key point of our approach is to model thin nets as the crest lines of the image surface. Crest lines are the lines where the magnitude of the maximum curvature is locally maximum in the corresponding principal direction. We define these lines using first, second and third derivatives of the image. We compute the image derivatives using recursive filters approximating the Gaussian filter and its derivatives. Using an adapted scale factor, we apply this approach to the extraction of roads in satellite data and blood vessels in medical images. We also apply this method to the extraction of the crest lines in depth maps of human faces.