期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2006
卷号：XXXVI Part 3
出版社：Copernicus Publications
摘要：A new methodology for the segmentation of imagery using network snakes is presented in this paper. Snakes are a well known technique, but usually are limited to closed object boundaries. Enhancing traditional snakes the focus is on objects forming a network respectively being adjacent with only one boundary in between. In addition, the focus is on linear objects with open non-fixed endings. The internal energy controlling the shape of the object contours during the energy minimization process is defined for nodes with different degrees to enable the exploitation of the object topology. Exemplary results of two different applications demonstrate the functionality and transferability of the proposed methodology: First, field boundaries are extracted from high resolution satellite imagery. The second example from the medical sector deals with the delineation of adjacent cells in microscopic cell imagery. Concluding remarks are given at the end to point out further investigations
