Skeletonizing Topographical Regions for Navigational Path Planning

This report investigates the application of two Computer Vision techniques for characterizing critical regions in digital terrain maps: Skeletonization and Topographical Region Detection. The Navigation Path Planning problem involves nding the optimal path through a digital terrain Map and must minimize a cost function while satisfying the constraints of navigation. This process is computationally expensive and has typically been intractable for in ight planning applications. The aim of this work is to extract representations of digital terrain maps which are relevant to the requirements of the path planning problem. Clustering and Di erential Geometry is used to detect regions which are critical to the cost function. Skeletonization is then used to extract features from these regions in terms of navigational constraints and provides a graph like representation. This allows the grid elements of digital terrain maps (pixels) to be replaced by critical regions (Skeletonized topographical regions) and substantially reduces the search required for in ight path planning.