Removal of water‐surface reflection effects with a temporal minimum filter for UAV‐based shallow‐water photogrammetry

The recent development of structure‐from‐motion (SfM) and multi‐view stereo (MVS) photogrammetry techniques has enabled semi‐automatic high‐resolution bathymetry using aerial images taken by consumer‐grade digital cameras mounted on unmanned aerial vehicles (UAVs). However, the applicability of these techniques is sometimes limited by sun and sky reflections at the water surface, which render the point‐cloud density and accuracy insufficient. In this research, we present a new imaging technique to suppress the effect of these water‐surface reflections. In this technique, we order a drone to take a short video instead of a still picture at each waypoint. We then apply a temporal minimum filter to the video. This filter extracts the smallest RGB values in all the video frames for each pixel, and composes an image with greatly reduced reflection effects. To assess the performance of this technique, we applied it at three small shallow‐water sites. Specifically, we evaluated the effect of the technique on the point cloud density and the accuracy and precision of the photogrammetry. The results showed that the proposed technique achieved a far denser point cloud than the case in which a randomly chosen frame was used for each waypoint, and also showed better overall accuracy and precision in estimating water‐bottom elevation. The effectiveness of this new technique should depend on the surface wave state and sky radiance distribution, and this dependence, as well as the applicability to large areas, should be investigated in future research. Copyright © 2018 John Wiley & Sons, Ltd.     

Digital mapping of coastal boulders – high-resolution data acquisition to infer past and recent transport dynamics

Coastal boulder fields provide clues to long-term frequency-magnitude patterns of coastal flooding events and have the potential to play an important role in coastal hazard assessment. Mapping boulders in the field is time and labour-intensive, and work on intertidal reef platforms, as in the present study, is physically challenging. By addressing coastal scientists who are not specialists in remote sensing, this contribution reports on the possibilities and limitations of digital applications in boulder mapping in Eastern Samar, Philippines, where recent supertyphoons Haiyan and Hagupit induced high waves, coastal flooding and boulder transport. It is demonstrated how satellite imagery of sub-metre resolution (from Pléiades and WorldView-3 imagery) enables efficient analysis of transport vectors and distances of larger boulders, reflecting variation in latitudes of both typhoon tracks and approaching angles of typhoon-generated waves. During the investigated events, boulders with a-axes of up to 8 m were clearly identified to have been shifted for up to 32 m, mostly along the seaward margin of the boulder field. It is, however, hard to keep track of smaller boulders, and the length of a-axes and b-axes including their orientation is often impossible to map with sufficient accuracy. Orthophotographs and digital surface models created through the application of an unmanned aerial vehicle and the ‘Structure from Motion’ technique provide ultra-high-resolution data, and have the potential to not only improve the results of satellite image analysis, but also those from field mapping and may significantly reduce overall time in the field. Orthophotographs permit unequivocal mapping of a-axes and b-axes including their orientation, while precise values for c-axes can be derived from the respective digital surface models. Volume of boulders is best inferred from boulder-specific Structure from Motion-based three-dimensional models. Battery power, flight speed and altitude determine the limits of the area covered, while patches shielded by the boulders are difficult to resolve. For some tasks, field mapping remains mandatory and cannot be replaced by currently available remote sensing tools: for example, sampling for rock type, density and age dating, recording of lithological separation of boulders from the underlying geological unit and of geomorphic features on a millimetre to decimetre-scale, or documentation of fine-grained sediment transport in between the boulders in supratidal settings. In terms of future events, the digital products presented here will provide a valuable reference to track boulder transport on a centimetre to decimetre-scale and to better understand the hydrodynamics of extreme-wave events on a fringing reef coastline.     

利用R语言半自动化提取河流阶地——以米家山黄河阶地为例

构造活跃地区阶地的发育对于分析不同时间域下的构造变形或气候作用具有重要意义。因此,如何利用有效的定量方法提取和精细刻画这类地貌特征显得极为重要。R语言是一种集统计分析和图形显示于一体的优秀编程语言,目前已被广泛应用于医学、生物学等领域,但尚未应用于地质与地貌学领域。文中以海原断裂带景泰—哈思山段的米家山东侧保存较好的多级黄河河流阶地为研究目标,初步尝试基于R语言对Sf M技术获取的高精度地形数据进行分析和可视化,完成了对米家山黄河阶地的半自动提取,共划分出20级河流阶地,同时揭示出较年轻的阶地具有较好的连续性和延伸性,而较老阶地的连续性和延伸性则相对较差,且老阶地变形逐渐趋于明显,阶地年龄越老,其似半抛物线形态的翘曲越明显,反映了米家山东侧多级阶地形成后的不同演化历史。此次试验结果表明R语言有望成为高精度地形数据分析和可视化的有效工具。     

Patterns and drivers of peat topographic changes determined from Structure-from-Motion photogrammetry at field plot and laboratory scales

Little is known about the spatial and temporal variability of peat erosion nor some of its topographic and weather-related drivers. We present field and laboratory observations of peat erosion using Structure-from-Motion (SfM) photogrammetry. Over a 12 month period, 11 repeated SfM surveys were conducted on four geomorphological sites of 18–28 m² (peat hagg, gully wall, riparian area and gully head) in a blanket peatland in northern England. A net topographic change of –14 to +30 mm yr^–1 for the four sites was observed during the whole monitoring period. Cold conditions in the winter of 2016 resulted in highly variable volume change (net surface topographic rise first and lowering afterwards) via freeze–thaw processes. Long periods of dry conditions in the summer of 2017 led to desiccation and drying and cracking of the peat surface and a corresponding surface lowering. Topographic changes were mainly observed over short-term intervals when intense rainfall, flow wash, needle-ice production or surface desiccation was observed. In the laboratory, we applied rainfall simulations on peat blocks and compared the peat losses quantified by traditional sediment flux measurements with SfM derived topographic data. The magnitude of topographic change determined by SfM (mean value: 0.7 mm, SD: 4.3 mm) was very different to the areal average determined by the sediment yield from the blocks (mean value: –0.1 mm, SD: 0.1 mm). Topographic controls on spatial patterns of topographic change were illustrated from both field and laboratory surveys. Roughness was positively correlated to positive topographic change and was negatively correlated to negative topographic change at field plot scale and laboratory macroscale. Overall, the importance of event-scale change and the direct relationship between surface roughness and the rate of topographic change are important characteristics which we suggest are generalizable to other environments. © 2018 John Wiley & Sons, Ltd.     

Assessing the performance of structure‐from‐motion photogrammetry and terrestrial LiDAR for reconstructing soil surface microtopography of naturally vegetated plots

Soil microtopography is a property of critical importance in many earth surface processes but is often difficult to quantify. Advances in computer vision technologies have made image‐based three‐dimensional (3D) reconstruction or Structure‐from‐Motion (SfM) available to many scientists as a low cost alternative to laser‐based systems such as terrestrial laser scanning (TLS). While the performance of SfM at acquiring soil surface microtopography has been extensively compared to that of TLS on bare surfaces, little is known about the impact of vegetation on reconstruction performance. This article evaluates the performance of SfM and TLS technologies at reconstructing soil microtopography on 6 m × 2 m erosion plots with vegetation cover ranging from 0% to 77%. Results show that soil surface occlusion by vegetation was more pronounced with TLS compared to SfM, a consequence of the single viewpoint laser scanning strategy adopted in this study. On the bare soil surface, elevation values estimated with SfM were within 5 mm of those from TLS although long distance deformations were observed with the former technology. As vegetation cover increased, agreement between SfM and TLS slightly degraded but was significantly affected beyond 53% of ground cover. Detailed semivariogram analysis on meter‐square‐scale surface patches showed that TLS and SfM surfaces were very similar even on highly vegetated plots but with fine scale details and the dynamic elevation range smoothed out with SfM. Errors in the TLS data were mainly caused by the distance measurement function of the instrument especially at the fringe of occlusion regions where the laser beam intersected foreground and background features simultaneously. From this study, we conclude that a realistic approach to digitizing soil surface microtopography in field conditions can be implemented by combining strengths of the image‐based method (simplicity and effectiveness at reconstructing soil surface under sparse vegetation) with the high accuracy of TLS‐like technologies. Copyright © 2015 John Wiley & Sons, Ltd.     

DATA COMPARATIVE ANALYSIS BETWEEN SFM DATA AND DGPS DATA: A CASE STUDY FROM FAULT SCARP IN THE EAST BANK OF HONGSHUIBA RIVER,NORTHERN MARGIN OF THE QILIAN SHAN

With the development of the techniques acquiring high-resolution digital terrain data,the digital terrain data acquisition technology has been widespread applied to the geoscience research.A revolutionary,low-cost and simply operative SfM (Structure from Motion) technology will make obtain high-resolution DEM data more convenient for researches on active tectonics.This paper summarizes the basic principles and workflows of SfM technology and processes and selects the Hongshuiba River area along the northern margin of the Qilian Shan to conduct data collection.We use a series of digital pictures to produce a texture with geographic information,in which data resolution is 6.73cm/pix and average density of point cloud is 220.667 point/m².The coverage area is 0.286km².Further,in order to compare the accuracy between SfM data and differential GPS (DGPS) data in details,SfM data are vertically shifted and tilt-corrected.After optimizing corrections of SfM data,the absolute value of elevation difference between two data substantially concentrates around 20cm,roughly equivalent to 2-folds of data error only after the elevation error correction.Elevation difference between two data is 10~15cm in 90% confidence interval.The maximum error is about 30cm,but accounts for less than 10%.Along the direction of fault trace,the height of fault scarp extracted from SfM data shows that vertical displacement of the latest tectonic activity in the east bank of Hongshuiba River is about 1m,and some minimum scarps height may be 0.3m.The results show SfM technology with high vertical accuracy can be able to replace differential GPS in high-precision topographic survey.After correcting of SfM data,elevation difference still exists,which may be associated with methods of generating DEM and SfM data accuracy,which in turn is controlled by the number and distribution of Ground Control Points (GCPs),photos density and camera shooting height,but also related to surface features,Fodongmiao-Hongyazi Fault     

Open-pit mining geomorphic feature characterisation

Among the anthropogenic topographic signatures on Earth, open-pit mines are of great importance. Mining is of interest to geomorphologists and environmental researchers because of its implication in geomorphic hazards and processes. In addition, open-pit mines and quarries are considered the most dangerous industrial sector, with injuries and accidents occurring in numerous countries. Their fast, accurate and low-cost investigation, therefore, represents a challenge for the Earth science community. The purpose of this work is to characterise the open-pit mining features using high-resolution topography and a recently published landscape metric, the Slope Local Length of Auto-Correlation (SLLAC) (Sofia et al., 2014). As novel steps, aside from the correlation length, the terrace’s orientation is also calculated, and a simple empirical model to derive the percentage of artificial surfaces is tested. The research focuses on two main case studies of iron mines, both located in the Beijing district (P.R. China). The main topographic information (Digital Surface Models, DSMs) was derived using an Unmanned Aerial Vehicle (UAV) and the Structure from Motion (SfM) photogrammetric technique. The results underline the effectiveness of the adopted methodologies and survey techniques in the characterisation of the main mine’s geomorphic features. Thanks to the SLLAC, the terraced area given by open-cast/open-pit mining for iron extraction is automatically depicted, thus, allowing researchers to quickly estimate the surface covered by the open-pit. This information could be used as a starting point for future research (i) given the availability of multi-temporal surveys to track the changes in the extent of the mine; (ii) to relate the extent of the mines to the amount of processes in the area (e.g. pollution, erosion, etc.), and to (iii) combine the two points, and analyse the effects of the change related to changes in erosion. The analysis of the correlation length orientation also allows researchers to identify the terrace’s orientation and to understand the shape of the open-pit area. The tectonic environment and history, or inheritance, of a given slope can determine if and how it fails, and the orientation of the topographic surface or excavation face, with respect to geologic features, is of major significance. Therefore, the proposed approach can provide a basis for a large-scale and low-cost topographic survey for sustainable environmental planning and, for example, for the mitigation of environmental anthropogenic impacts due to mining.     

APPLICATION OF SEMIAUTOMATIC EXTRACTION OF FLUVIAL TERRACES BASED ON R LANGUAGE-AN EXAMPLE FROM THE YELLOW RIVER TERRACES AT MIJIA SHAN

The generation, abandonment and preservation of terraces formed in active tectonic areas are important to the analysis of the role of the tectonics and climate along the temporal variations, so it appears significant as how to use the effective quantitative methods to extract and accurately depict these terraces. The increasingly convenient acquisition of high-precision topographic data has greatly promoted the advancement of quantitative research in geoscience, making it possible to analyze mid-micro-geomorphic features on a large scale, especially by studying the temporal and spatial evolution of tectonic deformation through accurate capture of micro-geomorphic features. Over the past decade, the rapid development of LiDAR(Light Detection and Ranging)technology has provided unprecedented opportunity to access high-precision topographic data(up to centimeter in vertical and horizontal directions). However, its relatively high cost and relatively complex data processing techniques limit its widespread application in the field of earth sciences. In recent years, with the continuous innovation and advancement of topographic measurement technology, the three-dimensional structure of motion reconstruction technology(Structure from Motion, SfM)has gradually been introduced into the field of digital topographic photogrammetry due to its rapid advantage in providing quick, convenient and cost-effective methods for obtaining high-density geospatial point data. This method thus shows great potential for providing high resolution topographic data with comparable resolution and precision. Therefore, with the acquisition of more and more high-resolution terrain data in recent years, it is an important development trend to explore automated or semi-automated quantitative geomorphological analysis methods. R language, as an excellent programming language, has not been used in the geology and geomorphology, although is widely applied in medicine and meteorology based on its powerful capability of statistician and graphic visualization. In this paper, we focus on the Yellow River multi-terraces formed to the east of the Mijia Shan, which belongs to the Jingtai-Hasi Shan segment of the Haiyuan Fault. With the analysis and visualization of the high-resolution topographic data collected from the SfM in the environment of the R language, we implement the semiautomatic classification and mapping of the Yellow River multi-terraces. The method identifies 20 terraces with different elevation. Our results also imply that the younger terraces have better continuity and elongation, and the older terraces have more deformation, which can be demonstrated from their gradually notable semi-parabolic shape. Besides this, it also suggests the diverse evolution stages of the Yellow River terraces. Our study indicates that R language is expected to become an efficient tool of statistics and visualization of the high-resolution topographic data.     

Archival photogrammetric analysis of river–floodplain systems using Structure from Motion (SfM) methods

In this study we evaluate the extent to which accurate topographic data can be obtained by applying Structure from Motion (SfM) photogrammetric methods to archival imagery. While SfM has proven valuable in photogrammetric applications using specially acquired imagery (e.g. from unmanned aerial vehicles), it also has the potential to improve the precision of topographic data and the ease with which can be produced from historical imagery. We evaluate the application of SfM to a relatively extreme case, one of low relative relief: a braided river–floodplain system. We compared the bundle adjustments of SfM and classical photogrammetric methods, applied to eight dates. The SfM approach resulted in data quality similar to the classical approach, although the lens parameter values (e.g. focal length) recovered in the SfM process were not necessarily the same as their calibrated equivalents. Analysis showed that image texture and image overlap/configuration were critical drivers in the tie‐point generation which impacted bundle adjustment quality. Working with archival imagery also illustrated the general need for the thorough understanding and careful application of (commercial) SfM software packages. As with classical methods, the propagation of (random) error in the estimation of lens and exterior orientation parameters using SfM methods may lead to inherent systematic error in the derived point clouds. We have shown that linear errors may be accounted for by point cloud registration based on a reference dataset, which is vital for the further application in quantitative morphological analyses when using archival imagery. Copyright © 2016 John Wiley & Sons, Ltd.