DepthLearn: Learning to Correct the Refraction on Point Clouds Derived from Aerial Imagery for Accurate Dense Shallow Water Bathymetry Based on SVMs-Fusion with LiDAR Point Clouds

被引:37
作者
Agrafiotis, Panagiotis [1 ,2 ]
Skarlatos, Dimitrios [2 ]
Georgopoulos, Andreas [1 ]
Karantzalos, Konstantinos [1 ]
机构
[1] Natl Tech Univ Athens, Dept Topog, Sch Rural & Surveying Engn, Zografou Campus,9 Heroon Polytech Str, Athens 15780, Greece
[2] Cyprus Univ Technol, Lab Photogrammetr Vis, Civil Engn & Geomat Dept, 2-8 Saripolou Str, CY-3036 Limassol, Cyprus
关键词
point cloud; bathymetry; SVM; machine learning; UAV; aerial imagery; seabed mapping; refraction effect; LiDAR; fusion; data integration; TOPOGRAPHY; ERRORS; MODEL;
D O I
10.3390/rs11192225
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
The determination of accurate bathymetric information is a key element for near offshore activities; hydrological studies, such as coastal engineering applications, sedimentary processes, hydrographic surveying, archaeological mapping and biological research. Through structure from motion (SfM) and multi-view-stereo (MVS) techniques, aerial imagery can provide a low-cost alternative compared to bathymetric LiDAR (Light Detection and Ranging) surveys, as it offers additional important visual information and higher spatial resolution. Nevertheless, water refraction poses significant challenges on depth determination. Till now, this problem has been addressed through customized image-based refraction correction algorithms or by modifying the collinearity equation. In this article, in order to overcome the water refraction errors in a massive and accurate way, we employ machine learning tools, which are able to learn the systematic underestimation of the estimated depths. In particular, an SVR (support vector regression) model was developed, based on known depth observations from bathymetric LiDAR surveys, which is able to accurately recover bathymetry from point clouds derived from SfM-MVS procedures. Experimental results and validation were based on datasets derived from different test-sites, and demonstrated the high potential of our approach. Moreover, we exploited the fusion of LiDAR and image-based point clouds towards addressing challenges of both modalities in problematic areas.
引用
收藏
页数:31
相关论文
共 74 条
[1]   SHALLOW WATER BATHYMETRY MAPPING FROM UAV IMAGERY BASED ON MACHINE LEARNING [J].
Agrafiotis, P. ;
Skarlatos, D. ;
Georgopoulos, A. ;
Karantzalos, K. .
UNDERWATER 3D RECORDING AND MODELLING: A TOOL FOR MODERN APPLICATIONS AND CH RECORDING, 2019, 42-2 (W10) :2-9
[2]  
Agrafiotis P, 2015, INT ARCH PHOTOGRAMM, V45, P1, DOI [10.5194/isprsarchives-XL-5-W5-1-2015, 10.5194/isprsarchives-XL-3-W2-1-2015]
[3]  
Agrafiotis P., 2018, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, VXLII2, P15, DOI DOI 10.5194/ISPRS-ARCHIVES-XLII-2-15-2018
[4]   Comparison of LiDAR waveform processing methods for very shallow water bathymetry using Raman, near-infrared and green signals [J].
Allouis, Tristan ;
Bailly, Jean-Stephane ;
Pastol, Yves ;
Le Roux, Catherine .
EARTH SURFACE PROCESSES AND LANDFORMS, 2010, 35 (06) :640-650
[5]   Merging Unmanned Aerial Systems (UAS) Imagery and Echo Soundings with an Adaptive Sampling Technique for Bathymetric Surveys [J].
Alvarez, Laura, V ;
Moreno, Hernan A. ;
Segales, Antonio R. ;
Pham, Tri G. ;
Pillar-Little, Elizabeth A. ;
Chilson, Phillip B. .
REMOTE SENSING, 2018, 10 (09)
[6]  
[Anonymous], P 1 INT C APPL PHYS
[7]  
[Anonymous], PHOTOGRAPHY ARCHAELO
[8]  
[Anonymous], 1963, Photogrammetric Eng
[9]  
[Anonymous], INT ARCH PHOTOGRA B5
[10]  
[Anonymous], P 20 AUSTR SURV C DA