Machine Vision and Applications (2016) 27:87–101 DOI 10.1007/s00138-015-0723-9 ORIGINAL PAPER Characterization of mountain drainage patterns for GPS-denied UAS navigation augmentation Teng Wang 1 · Koray Celik 1 · Arun K. Somani 1 Received: 16 December 2014 / Revised: 1 June 2015 / Accepted: 15 September 2015 / Published online: 26 October 2015 © Springer-Verlag Berlin Heidelberg 2015 Abstract We present a novel approach to use moun- tain drainage patterns for GPS-denied navigation of small unmanned aerial systems, such as the ScanEagle, utiliz- ing a down-looking fixed focus monocular imager. Our proposal allows for extension of GPS-denied missions in mountain areas, with no assumption of human-made geo- graphic objects. We leverage the analogy between mountain drainage patterns, human aeteriograms, and human finger- prints, to match local drainage patterns to parallax occlusion maps of geo-registered radar returns (GRRR). The GRRR data may be loaded on-board the aircraft pre-mission to avoid the need for a scanning aperture radar during the mission. For recognition purpose, we represent a given mountain area with a set of spatially distributed mountain minutiae, i.e., details found in the drainage patterns, so that conventional minutiae-based fingerprint matching approaches can be used to match real-time camera images against template images in the training set. We use medical arteriography processing technique to extract the patterns. The minutiae-based repre- sentation of mountains is achieved by first exposing mountain ridges and valleys with a series of filters and then extract- ing mountain minutiae from these ridges/valleys. Our results are experimentally validated on actual terrain data and show the effectiveness of minutiae-based mountain representation method. B Teng Wang tengw@iastate.edu Koray Celik koray@iastate.edu Arun K. Somani arun@iastate.edu 1 Department of Electrical and Computer Engineering, Iowa State University, Ames, IA 50010, USA Keywords UAS navigation · GPS-denied terrain area · Mountain fingerprints · Drainage patterns · Minutiae · GIS 1 Introduction Unmanned aircraft systems (UASs) offer fascinating research areas. An UAS is defined as an aircraft with no on-board human pilot. The use of UASs have been attracting much attention among geometrics and geospatial professionals recently. The applications of such UASs are diverse, ranging from scientific exploration and data collection, to provision of commercial services, military reconnaissance, and intel- ligence gathering. Other areas include law enforcement, and even entertainment. Currently, most UASs are drones which rely on a substan- tial degree of human remote control. Recent versions have built-in control and/or guidance systems to perform speed and/or flight path stabilization and simple prescribed naviga- tion functions such as way-point following [2]. For an UAS to replace the human pilot and fly in all kinds of environ- ments, the current state of the art in UAS autonomy falls short of the necessary capabilities. Global positioning sys- tem (GPS) is a critical sensor for UAS navigation due to its accuracy, global coverage, and small hardware footprint. However, GPS is subject to interruption or denial due to sig- nal blockage or RF interference, such as through canyons or under forest canopy. When GPS is unavailable, position, velocity and altitude (PVA) performance from other iner- tial and air data sensors is not sufficient, especially for small UASs. Therefore, a key issue of UAS design is how to enable them to operate during GPS denial. To overcome the problem, cutting-edge system designs are specified using concepts of hybrid navigation, fusing in real- time all available navigation data, such as radar altimeters, 123