Variable shape models for LS-based automatic building extraction from VHR satellite imagery

Abstract

In this paper, we propose a level set based automatic building extraction method using prior shapes. We introduce a variable shape model which together with the level set function for segmentation dynamically indicates the region with which the prior shape should be compared. Our model is capable of segmenting an object from an image based on the image intensity as well as the prior shape. In addition, the proposed model permits translation, scaling and rotation of the prior shape. Moreover, a fast way is also established for the minimization of our functional. The experiments validate our model.