  Citation: Qin, R.; Ling, X.; Farella, E.M.; Remondino, F. Uncertainty-Guided Depth Fusion from Multi-View Satellite Images to Improve the Accuracy in Large-Scale DSM Generation. Remote Sens. 2022, 14, 1309. https://doi.org/10.3390/ rs14061309 Academic Editors: Wojciech Drzewiecki, Beata Hejmanowska and Slawomir Mikrut Received: 21 January 2022 Accepted: 3 March 2022 Published: 8 March 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affil- iations. Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). remote sensing Article Uncertainty-Guided Depth Fusion from Multi-View Satellite Images to Improve the Accuracy in Large-Scale DSM Generation Rongjun Qin 1,2,3,4, * , Xiao Ling 1,2 , Elisa Mariarosaria Farella 5 and Fabio Remondino 5 1 Geospatial Data Analytics Laboratory, The Ohio State University, 218B Bolz Hall, 2036 Neil Avenue, Columbus, OH 43210, USA; xlingsky@gmail.com 2 Department of Civil, Environmental and Geodetic Engineering, The Ohio State University, 218B Bolz Hall, 2036 Neil Avenue, Columbus, OH 43210, USA 3 Department of Electrical and Computer Engineering, The Ohio State University, 2036 Neil Avenue, Columbus, OH 43210, USA 4 Translational Data Analytics Institute, The Ohio State University, 1760 Neil Avenue, Columbus, OH 43210, USA 5 3D Optical Metrology Unit, Bruno Kessler Foundation (FBK), Via Sommarive 18, 38123 Trento, Italy; elifarella@fbk.eu (E.M.F.); remondino@fbk.eu (F.R.) * Correspondence: qin.324@osu.edu; Tel.: +1-614-292-4356 Abstract: The generation of digital surface models (DSMs) from multi-view high-resolution (VHR) satellite imagery has recently received a great attention due to the increasing availability of such space-based datasets. Existing production-level pipelines primarily adopt a multi-view stereo (MVS) paradigm, which exploit the statistical depth fusion of multiple DSMs generated from individual stereo pairs. To make this process scalable, these depth fusion methods often adopt simple ap- proaches such as the median filter or its variants, which are efficient in computation but lack the flexibility to adapt to heterogenous information of individual pixels. These simple fusion approaches generally discard ancillary information produced by MVS algorithms (such as measurement con- fidence/uncertainty) that is otherwise extremely useful to enable adaptive fusion. To make use of such information, this paper proposes an efficient and scalable approach that incorporates the matching uncertainty to adaptively guide the fusion process. This seemingly straightforward idea has a higher-level advantage: first, the uncertainty information is obtained from global/semiglobal matching methods, which inherently populate global information of the scene, making the fusion process nonlocal. Secondly, these globally determined uncertainties are operated locally to achieve efficiency for processing large-sized images, making the method extremely practical to implement. The proposed method can exploit results from stereo pairs with small intersection angles to recover details for areas where dense buildings and narrow streets exist, but also to benefit from highly accurate 3D points generated in flat regions under large intersection angles. The proposed method was applied to DSMs generated from Worldview, GeoEye, and Pleiades stereo pairs covering a large area (400 km 2 ). Experiments showed that we achieved an RMSE (root-mean-squared error) improvement of approximately 0.1–0.2 m over a typical Median Filter approach for fusion (equivalent to 5–10% of relative accuracy improvement). Keywords: satellite photogrammetry; multi-view stereo; depth fusion; digital surface models; dense image matching; uncertainty 1. Introduction The number of very high-resolution (VHR) optical satellite sensors has increased drastically over the last two decades. These sensors, such as WorldView I-IV, Pleiades A/B, PleiadesNeo, SkySat, GaoFen, KompSat, etc. [1], are capable of collecting images at a resolution of one meter or less with large swaths, adding petabytes of data to the archives Remote Sens. 2022, 14, 1309. https://doi.org/10.3390/rs14061309 https://www.mdpi.com/journal/remotesensing