Engineering Department, University of Ferrara, Saragat 1, 44122 Ferrara, Italy.
Physics and Earth Science Department, University of Ferrara, Saragat 1, 44122 Ferrara, Italy.
Sensors (Basel). 2019 Apr 10;19(7):1717. doi: 10.3390/s19071717.
Coastal environments are usually characterized by a brittle balance, especially in terms of sediment transportation. The formation of dunes, as well as their sudden destruction as a result of violent storms, affects this balance in a significant way. Moreover, the growth of vegetation on the top of the dunes strongly influences the consequent growth of the dunes themselves. This work presents the results obtained through a long-term monitoring of a complex dune system by the use of Unmanned Aerial Vehicles (UAVs). Six different surveys were carried out between November 2015 and December 2017 in the littoral of Rosolina Mare (Italy). Aerial photogrammetric data were acquired during flight repetitions by using a DJI Phantom 3 Professional with the camera in a nadiral arrangement. The processing of the captured images consisted of the reconstruction of a three-dimensional model using the Structure-from-Motion (SfM). Each model was framed in the European Terrestrial Reference System (ETRS) using GNSS geodetic receivers in Network Real Time Kinematic (NRTK). Specific data management was necessary due to the vegetation by filtering the dense cloud. This task was performed by both performing a slope detection and a removal of the residual outliers. The final products of this approach were thus represented by Digital Elevation Models (DEMs) of the sandy coastal section. In addition, DEMs of Difference (DoD) were also computed for the purpose of monitoring over time and detecting variations. The accuracy assessment of the DEMs was carried out by an elevation comparison through especially GNSS-surveyed points. Relevant cross sections were also extracted and compared. The use of the Structure-from-Motion approach by UAVs finally proved to be both reliable and time-saving thanks to quicker in situ operations for the data acquisition and an accurate reconstruction of high-resolution elevation models. The low cost of the system and its flexibility represent additional strengths, making this technique highly competitive with traditional ones.
沿海环境通常具有脆弱的平衡特性，特别是在泥沙输运方面。沙丘的形成以及由于剧烈风暴导致的突然破坏，对这种平衡产生了重大影响。此外，沙丘顶部植被的生长对沙丘本身的后续生长有很大的影响。本工作通过使用无人机（UAV）对一个复杂沙丘系统进行长期监测，展示了所获得的结果。在 2015 年 11 月至 2017 年 12 月期间，在意大利罗索利纳马雷（Rosolina Mare）的滨海地区进行了六次不同的调查。在使用 DJI Phantom 3 Professional 相机进行的重复飞行中获取了航空摄影测量数据，相机采用天底方向排列。捕获的图像的处理包括使用基于运动结构（SfM）的方法重建三维模型。每个模型都使用 GNSS 大地测量接收机在网络实时动态（NRTK）中纳入欧洲大地参考系（ETRS）。由于植被的存在，需要进行特定的数据管理，以过滤密集云。这项任务通过执行坡度检测和去除残余异常值来完成。该方法的最终产品是用数字高程模型（DEM）表示的沙质沿海部分。此外，还计算了用于随时间监测和检测变化的差值数字高程模型（DoD）。DEM 的精度评估是通过特别使用 GNSS 测量点进行高程比较来完成的。还提取并比较了相关的横截面。最终证明，无人机的运动结构方法既可靠又节省时间，因为数据采集的现场操作更快，并且可以准确重建高分辨率高程模型。该系统的低成本和灵活性是其额外的优势，使其技术具有很强的竞争力。
