Remote sensing image object extraction using convex geometric active contour model
Ning He, Lulu Zhang, Yixue Wang · 2013
The geometric active contour model is a popular method for computing the segmentation of an image into two phases, based on Mumford-Shah model. The main problem in image segmentation based this method may lead to non-convex minimization problems that it difficult to obtain a global solution. In this paper, we propose a convex relaxation of the popular K-means algorithm. Our approach is based on the vector-valued relaxation technique developed by Brown et al. (UCLA CAM Report 10-43, 2010) and Goldstein et al. (UCLA CAM Report 09-77, 2009). We applied the proposed framework to multi-object extraction problems on remote sensing images. We provide several experimental results to demonstrate that our convex model yields global solutions to the well known Mumford-Shah model.