Simultaneous Mapping of Coastal Topography and Bathymetry From a Lightweight Multicamera UAS

Unmanned aerial vehicle (UAV)-based snow depth is mapped as the difference between snow-on and snow-off digital surface models (DSMs), which are derived using the structure from motion (SfM) technique with ground control points (GCPs). In this study, we evaluated the impacts of the quality and deployment of GCPs on the accuracy of snow depth estimates. For 15 GCPs in our study area, we surveyed each of their coordinates using an ordinary global positioning system (GPS) and a differential GPS, producing two sets of GCP measurements (hereinafter, the low-accuracy and high-accuracy sets). The two sets of GCP measurements were then incorporated into SfM processing of UAV images by following two deployment strategies to create snow-off and snow-on DSMs and then to retrieve snow depth. In Strategy A, the same GCP measurements in each set were used to create both the snow-on and snow-off DSMs. In Strategy B, each set of GCP measurements was divided into two sub-groups, one sub-group for creating snow-on DSMs and the other sub-group for snow-off DSMs. The accuracy of snow depth estimates was evaluated in comparison to concurrent in-situ snow depth measurements. The results showed that Strategy A, using both the low-accuracy and high-accuracy sets, generated accurate snow depth estimates, while in Strategy B, only the high-accuracy set could generate reliable snow depth estimates. The results demonstrated that the deployment of GCPs had a significant influence on UAV-based SfM snow depth retrieval. When accurate GCP measurements cannot be guaranteed (e.g., in mountainous regions), Strategy A is the optimal option for producing reliable snow depth estimates. When highly accurate GCP measurements are available (e.g., collected by differential GPS in open space), both deployment strategies can produce accurate snow depth estimates.

Excessive streambank erosion is a significant source of fine sediments and associated nutrients in many river systems as well as poses risk to infrastructure. Geomorphic change detection using high-resolution topographic data is a useful method for monitoring the extent of bank erosion along river corridors. Recent advances in an unmanned aircraft system (UAS) and structure from motion (SfM) photogrammetry techniques allow acquisition of high-resolution topographic data, which are the methods used in this study. To evaluate the effectiveness of UAS-based photogrammetry for monitoring bank erosion, a fixed-wing UAS was deployed to survey 20 km of river corridors in central Vermont, in the northeastern United States multiple times over a two-year period. Digital elevation models (DEMs) and DEMs of difference allowed quantification of volumetric changes along selected portions of the survey area where notable erosion occurred. Results showed that UAS was capable of collecting high-quality topographic data at fine resolutions even along vegetated river corridors provided that the surveys were conducted in early spring, after snowmelt but prior to summer vegetation growth. Longer term estimates of streambank movements using the UAS showed good comparison to previously collected airborne lidar surveys and allowed reliable quantification of significant geomorphic changes along rivers.

Unmanned Aircraft Systems (UASs) can collect high resolution and high quality images for local mapping. Before the UAS images can be used for accurate mapping tasks in local areas, the precise position and orientation of the UAS images should first be determined. Direct georeferencing by POS (Position and Orientation System), a combination of GPS (Global Positioning System) and IMU (Inertial Measurement Unit), is the best choice; however, most commercial UASs cannot carry highly accurate IMUs because of the limited payload. Therefore, this study will discuss the accuracy of indirect georeferencing for UAS images. One approach for indirect georeferencing is general aerial triangulation (AT) by using well-distributed ground control points (GCPs). The other one is GPS-supported AT with GPS observations as airborne controls. In this paper, the camera is calibrated by the field method, and the accuracy of these two approaches for indirect georeferencing is presented. Based on 20 horizontal check points and 29 vertical check points, this study shows the stereoscopic viewing accuracy of general AT for UAS images, collected by Canon EOS 5D Mark II camera with 24 mm F/1.4L II USM lens at a flying height of 550 m, is about 0.26 m (ca. 1.73 pixels) in planimetry and 0.27 m (ca. 1.80 pixels) in height. GPS-supported AT produced the stereoscopic viewing accuracy about 0.44 m (ca. 2.93 pixels) in planimetry and 0.55 m (ca. 3.67 pixels) in height. The test results show that the accuracy of these two indirect georeferencing approaches of fixed-wing UAS images can be used for updating local 1/5,000 topographic maps in Taiwan.

Structure from motion (SfM) and MultiView Stereo (MVS) algorithms are increasingly being applied to imagery from unmanned aircraft systems (UAS) to generate point cloud data for various surveying and mapping applications. To date, the options for assessing the spatial accuracy of the SfM-MVS point clouds have primarily been limited to empirical accuracy assessments, which involve comparisons against reference data sets, which are both independent and of higher accuracy than the data they are being used to test. The acquisition of these reference data sets can be expensive, time consuming, and logistically challenging. Furthermore, these experiments are also almost always unable to be perfectly replicated and can contain numerous confounding variables, such as sun angle, cloud cover, wind, movement of objects in the scene, and camera thermal noise, to name a few. The combination of these factors leads to a situation in which robust, repeatable experiments are cost prohibitive, and the experiment results are frequently site-specific and condition-specific. Here, we present a workflow to render computer generated imagery using a virtual environment which can mimic the independent variables that would be experienced in a real-world UAS imagery acquisition scenario. The resultant modular workflow utilizes Blender, an open source computer graphics software, for the generation of photogrammetrically-accurate imagery suitable for SfM processing, with explicit control of camera interior orientation, exterior orientation, texture of objects in the scene, placement of objects in the scene, and ground control point (GCP) accuracy. The challenges and steps required to validate the photogrammetric accuracy of computer generated imagery are discussed, and an example experiment assessing accuracy of an SfM derived point cloud from imagery rendered using a computer graphics workflow is presented. The proposed workflow shows promise as a useful tool for sensitivity analysis and SfM-MVS experimentation.

Precisely measuring the Earth’s changing surface on decadal to centennial time scales is critical for many science and engineering applications, yet long-term records of quantitative landscape change are often temporally and geographically sparse. Archives of scanned historical aerial photographs provide an opportunity to augment these records with accurate elevation measurements that capture the historical state of the Earth surface. Structure from Motion (SfM) photogrammetry workflows produce high-quality digital elevation models (DEMs) and orthoimage mosaics from these historical images, but time-intensive tasks like manual image preprocessing (e.g., fiducial marker identification) and ground control point (GCP) selection impede processing at scale. We developed an automated method to process historical images and generate self-consistent time series of high-resolution (0.5–2 m) DEMs and orthomosaics, without manual GCP selection. The method relies on SfM to correct camera interior and exterior orientation and a robust multi-stage co-registration approach using modern reference terrain datasets for geolocation refinement. We demonstrate the method using scanned images from the North American Glacier Aerial Photography (NAGAP) archive collected between 1967 and 1997. We present results for two sites with variable photo acquisition geometry and overlap — Mount Baker and South Cascade Glacier in Washington State, USA. The automated method corrects initial camera position errors of several kilometers and produces accurately georeferenced, high-resolution DEMs and orthoimages, regardless of camera configuration, acquisition geometry, terrain characteristics, and reference DEM properties. The average RMS reprojection error following bundle adjustment optimization was 0.67 px (0.15 m) for the 261 images contributing to 10 final DEM mosaics between 1970 and 1992 at Mount Baker, and 0.65 px (0.13 m) for the 243 images contributing to 18 individual DEMs between 1967 and 1997 at South Cascade Glacier. The relative accuracy of elevation values in the historical time series stacks was 0.68 m at Mount Baker and 0.37 m at South Cascade Glacier. Our products have reduced systematic error and improved accuracy compared to DEM products generated using SfM with manual GCP selection. Final elevation change measurement precision was ∼0.7–1.0 m over a 30-year period, enabling the study of processes with rates as low as ∼1-3 cm/yr. Our results demonstrate the potential of this scalable method to rapidly process archives of historical imagery and deliver new quantitative insights on long-term geodetic change and Earth surface processes.

Lever-arm and boresight correction, and field of view determination of a spectroradiometer mounted on an unmanned aircraft system

Abstract. Landscapes have been photographed dozens of times at scales ca. 1/25,000 and better since World War II. Scans are distributed freely online (e.g. remonterletemps.ign.fr). In parallel, Structure-from-Motion (SFM) software made photogrammetric processing easy to non-specialists. Yet puzzling questions crop up to use both: (i) Can raw scans be used as is? (ii) Can Ground Control Points (GCP) and checkpoints be safely collected from a web portal? (iii) How many parameters are sufficient for camera interior orientation? (iv) Are single flight camera networks sufficient to constrain camera models compared to multiple flights? (v) Are photogrammetric Digital Surface Models (DSM) fit for quantifying landslide activity? Processing of scanned black-and-white 1/27,000 photographs from IGN flown in May 1978 over Cirque de Salazie in La Réunion Island answer these questions. We find that raw scanned photographs need translation, rotation and cropping to match the camera reference frame. GCP and Check point coordinates collected on geoportail.gouv.fr with assumed accuracy of 10 m, achieved ca. 7 m accurate SFM registration. The optimal camera model uses only 4 parameters: f, cx, cy and K1. Compared to a 2015 lidar Digital Terrain Model (DTM), the 0.66 m/pixel DSM of 1978 has a median deviation of −1.39 m ± 3.34 m (Median Absolute Deviation) which is comparable to GCP quality. Elevation difference more importantly reveals, for the first time, the 37 years and 13 cyclones cumulated landslides pattern on Cirque de Salazie. Photographic archives hold decades-long 3D history. SFM is a game changer for landslide risk mitigation planning.

Digital surface models (DSMs) extracted from high-resolution Radarsat-2 stereo-images using different geometric modeling (deterministic, new hybrid, and empirical) are evaluated. The 3-D deterministic models are Toutin's and hybrid Toutin's models (TM and HTM) developed at the Canada Centre for Remote Sensing, and the empirical model is the rational function model (RFM). TM is computed with one and eight ground control points (GCPs), HTM without GCP and RFM supplied by MacDonald, Dettwiler and Associates Ltd. is postprocessed with 3-9 GCPs depending of degrees of 2-D polynomial functions. The DSMs are then generated and compared to 0.2-m accurate lidar elevation data. Because DSMs included the height of land covers, elevation linear errors with 68% and 90% confidence level (LE68 and LE90) are computed and compared over bare surfaces only. LE90 results are: TM with eight GCPs achieves the best results (6.3 m), then HTM with no GCP (7 m), TM with one GCP (8.6 m), and finally RFM the worst (9.7 m) whatever the polynomial degree and GCP number. HTM is the only modeling not using any GCP, which offers a strong advantage in operational environments.

Interferometric synthetic aperture radar (InSAR) processing, if lacking high quality ground control points (GCPs), may produce large errors in the final DEM. Some kind of alignment or registration can reduce these errors. This paper evaluates the accuracy of InSAR processed digital elevation models (DEM), against a high resolution DEM. InSAR DEMs were aligned and least-squares registered with USGS, 1/3 arc second NED. The accuracy evaluations after registration show InSAR DEMs alignment and registration can eliminate DEM errors caused by lack of accurate GCPs. In some cases, this process improves the accuracy of the InSAR DEM even more efficiently than including GCPs into InSAR processing.

A method for ground control point (GCP) acquisition using a simulated synthetic aperture radar (SAR) image derived from a digital elevation model (DEM) is proposed. Also proposed is a method for the evaluation of the accuracy of GCP matching with texture features from a reference GCP chip image. Results from experiments with simulated GCP chip images as reference images and geometrically distorted GCP chip images, derived using simulated SAR images as current images, show good coincidence with GCP matching accuracy in terms of pixel distances between matched GCP points in reference and current chip images and texture features. In particular, the correlation coefficient between the angular second moment (ASM) and the matching accuracy is 0.757, followed by Ent (entropy), Horn (homoginiety), Dis (dissimilarity), Con (contrast), Var (variability coefficient) and Chi (Chi-square), for the skewed images while for the rotated images, Dis shows the highest correlation of 0.628, followed by Horn, Asm, Co...

Recent technological innovations have led to the available of lightweight Unmanned Aerial Vehicle (UAV) and Structure-from-Motion (SfM) photogrammetry that are successfully applied for 3D topographic surveys. However, application of UAV and SfM for complex topographic areas i.e. open-pit mine areas is still poorly understood. This paper aims to investigate and verify potential application of these techniques for generating Digital Surface Model (DSM) at open-pit coal mine area and assessing its accuracy. For this purpose, the Nui Beo open-pit coal mine located in northeast Vietnam is selected as a case study. Accordingly, a total of 206 photos were captured using DJI Phantom 3 Professional. In addition, 19 ground control points (GCPs) were established using Leica TS09 total station. The accuracy of DSM was assessed using root-mean-square error (RMSE) in X, Y, Z, XY, and XYZ components. The result showed that the DSM model has high accuracy, RMSE on the 12 calibrated GCPs for X, Y, Z, XY, XYZ is 1.1 cm, 1.9 cm, 0.8 cm, 2.2 cm, and 2.3 cm, respectively, whereas RMSE on the 7 checked GCPs is 1.8 cm, 2.4 cm, 3.2 cm, 3.0 cm, and 4.4 cm for X, Y, Z, XY, XYZ components, respectively. We concluded that small UAV and SfM are feasible and valid tools for 3D topographic mapping in complex terrains such as open-pit coal mine areas.

The inclusion of thermal infrared (TIR) data in point clouds derived from unmanned aircraft system (UAS) imagery can benefit a variety of applications in which surface temperature and 3D geometry are both important discriminants of feature type and condition. Low resolution and narrow fields of view (FOV) of current consumer-grade TIR cameras on UAS, combined with the lack of sharpness and texture in many image regions, may cause failure or poor results from structure from motion (SfM) photogrammetric software, which has gained widespread use for generating point clouds from UAS imagery. This paper proposes a photogrammetric approach for generating 3D multispectral point clouds utilizing coacquired TIR-RGB images. A 3D point cloud is first generated from the RGB imagery using standard SfM procedures. Then the TIR attributes are assigned to points, where the image coordinates of the points in TIR images are estimated using transformation parameters obtained from co-registration procedures. To obtain RGB-to-TIR transformation parameters, this study tests 3D and 2D co-registration approaches. The latter produces better results due to the challenge of calibrating the TIR camera as required for the 3D approach. This proposed approach is advantageous for generating TIR point clouds without loss of photogrammetric precision compared with solely TIR-based SfM, as the 3D accuracy, point density, and reliability are greatly enhanced.

Bathymetric mapping is an important tool for reservoir management, typically completed before reservoir construction. Historically, bathymetric maps were produced by interpolating between points measured at a relatively large spacing throughout a reservoir, typically on the order of a few, up to 10, meters or more depending on the size of the reservoir. These measurements were made using traditional survey methods before the reservoir was filled, or using sonar surveys after filling. Post-construction issues such as sedimentation and erosion can change a reservoir, but generating updated bathymetric maps is difficult as the areas of interest are typically in the sediment deltas and other difficult-to-access areas that are often above water or exposed for part of the year. We present a method to create complete reservoir bathymetric maps, including areas above the water line, using small unmanned aerial vehicle (sUAV) photogrammetry combined with multi-beam sonar data—both established methods for producing topographic models. This is a unique problem because the shoreline topographic models generated by the photogrammetry are long and thin, not an optimal geometry for model creation, and most images contain water, which causes issues with image-matching algorithms. This paper presents methods to create accurate above-water shoreline models using images from sUAVs, processed using a commercial software package and a method to accurately knit sonar and Structure from Motion (SfM) data sets by matching slopes. The models generated by both approaches are point clouds, which consist of points representing the ground surface in three-dimensional space. Generating models from sUAV-captured images requires ground control points (GCPs), i.e., points with a known location, to anchor model creation. For this study, we explored issues with ground control spacing, masking water regions (or omitting water regions) in the images, using no GCPs, and incorrectly tagging a GCP. To quantify the effect these issues had on model accuracy, we computed the difference between generated clouds and a reference point cloud to determine the point cloud error. We found that the time required to place GCPs was significantly more than the time required to capture images, so optimizing GCP density is important. To generate long, thin shoreline models, we found that GCPs with a ~1.5-km (~1-mile) spacing along a shoreline are sufficient to generate useful data. This spacing resulted in an average error of 5.5 cm compared to a reference cloud that was generated using ~0.5-km (~1/4-mile) GCP spacing. We found that we needed to mask water and areas related to distant regions and sky in images used for model creation. This is because water, objects in the far oblique distance, and sky confuse the algorithms that match points among images. If we did not mask the images, the resulting models had errors of more than 20 m. Our sonar point clouds, while self-consistent, were not accurately georeferenced, which is typical for most reservoir surveys. We demonstrate a method using cross-sections of the transition between the above-water clouds and sonar clouds to geo-locate the sonar data and accurately knit the two data sets. Shore line topography models (long and thin) and integration of sonar and drone data is a niche area that leverages current advances in data collection and processing. Our work will help researchers and practitioners use these advances to generate accurate post-construction reservoir bathometry maps to assist with reservoir management.

Though unmanned aircraft systems (UAS) are widely used in agriculture, their current positioning accuracy in a radius of 0.5 to 2 meters is still too low to pinpoint a crop row or to precisely overlay temporal multi-source field maps together without a valid geometric calibration. The positioning accuracy of UAS deployed with real time kinematic (RTK) global navigation satellite system (GNSS) can be largely increased to a centimeter level, which was claimed from the manufacturers. This paper includes the preliminary test results of positioning accuracy of a commercial RTK UAS over a set of fixed position panels in our customized scenarios. Images were collected in three GNSS modes (regular GNSS without RTK, RTK mode 1 - not corrected by the positioning error of the base station, and RTK mode 2 - corrected by the positioning error of the base station) in static and in-flight settings. In the static setting, horizontal accuracies were 2.17 cm for the RTK mode 2, 12.11 cm for the RTK mode 1, and 11.46 cm for the regular GNSS mode. The significant result of horizontal accuracy in the in-flight setting was that RTK mode 2 without GCPs (2.82 cm) showed comparable accuracy with the commonly used regular GNSS mode with GCPs (1.34 cm). The vertical positioning accuracy in the static setting were 6.01 cm for the RTK mode 2, 5.65 cm for the RTK mode 1, and 10.48 cm for the regular GNSS mode. The accuracy of height measurement from digital surface models (DSMs) without and with GCPs in RTK mode 2 were 4.81 cm and 3.72 cm, respectively, which were the best performance among the three modes. In summary, the RTK UAS tested in this study showed great potential in eliminating the requirement of using GCPs and in high-positioning-accuracy application. The next phase is to test the system in field for accurate crop height measurement at different growth stages in agricultural application.

Abstract. Vendor-provided rational polynomial coefficients (RPCs) are commonly used to generate digital elevation models (DEMs) from high-resolution satellite images. This results in a level of accuracy that can be improved using ground control points (GCPs). It is well known that due to the inherent bias of sensor orientation the generated DEM is distorted. In the traditional way of working, the bias is corrected by integrating GCPs into the standard processing chain. This involves additional effort, since the provision of GCPs and the measurement of their image coordinates are required.In this paper, we examine whether and how the data recorded by NASA's ICESat (Ice, Cloud, and Land Elevation Satellite) mission can be used as GCPs without measuring image coordinates. The starting point are DEMs that are generated by image matching from KOMPSAT-3 satellite images with given RPCs. We developed a point-to-surface matching method that matches the ICESat points to the DEM in order to correct the DEM and improve its precision. For the experimental investigations a KOMPSAT 3 image pair is used that covers an area of 20 by 16 km2 in the Yangsan city regions. The generated DEM has a height accuracy of about 9 m. The point-to-surface algorithm with 505 ICESat points led to an improvement of the DEM height accuracy to about 2 m.