Skeleton Extraction from Incomplete Boundaries in Sensor Networks Based on Distance Transform

Abstract

We study the problem of skeleton extraction for large-scale sensor networks using only connectivity information. Existing solutions for this problem heavily depend on an algorithm that can accurately detect network boundaries. This dependence may seriously affect the effectiveness of skeleton extraction. For example, in low density networks, boundary detection algorithms normally do not work well, potentially leading to an incorrect skeleton being generated. This paper proposes a novel approach, named DIST, to skeleton extraction from incomplete boundaries using the idea of distance transform, a concept in the computer graphics area. The main contribution is a distributed and low-cost algorithm that produces accurate network skeletons without requiring that the boundaries be complete or tight. The algorithm first establishes the network's distance transform - the hop distance of each node to the network's boundaries. Based on this, some critical skeleton nodes are identified. Next, a set of skeleton arcs are generated by controlled flooding; connecting these skeleton arcs then gives us a coarse skeleton. The algorithm finally refines the coarse skeleton by building shortest path trees, followed by a prune phase. The obtained skeletons are robust to boundary noise and shape variations.