Abstract
Abstract Areal deformation monitoring based on point clouds can be a very valuable alternative to the established point-based monitoring techniques, especially for deformation monitoring of natural scenes. However, established deformation analysis approaches for point clouds do not necessarily expose the true 3D changes, because the correspondence between points is typically established naïvely. Recently, approaches to establish the correspondences in the feature space by using local feature descriptors that analyze the geometric peculiarities in the neighborhood of the interest points were proposed. However, the resulting correspondences are noisy and contain a large number of outliers. This impairs the direct applicability of these approaches for deformation monitoring. In this work, we propose Feature to Feature Supervoxel-based Spatial Smoothing (F2S3), a new deformation analysis method for point cloud data. In F2S3 we extend the recently proposed feature-based algorithms with a neural network based outlier detection, capable of classifying the putative pointwise correspondences into inliers and outliers based on the local context extracted from the supervoxels. We demonstrate the proposed method on two data sets, including a real case data set of a landslide located in the Swiss Alps. We show that while the traditional approaches, in this case, greatly underestimate the magnitude of the displacements, our method can correctly estimate the true 3D displacement vectors.
Highlights
Areal deformation monitoring based on point clouds can be a very valuable alternative to the established point-based monitoring techniques, especially for deformation monitoring of natural scenes
In F2S3 we extend the recently proposed feature-based algorithms with a neural network based outlier detection, capable of classifying the putative pointwise correspondences into inliers and outliers based on the local context extracted from the supervoxels
F2S3 consists of three main modules (i) a local feature descriptor (3DSmoothNet) used to infer the feature vectors of all points in the point clouds of both epochs (ii) a novel neural network (NN) based outlier detection algorithm used to robustify the initial set of correspondences established in the feature space and (iii) a supervoxel segmentation algorithm that provides the boundaries for the spatial smoothing
Summary
Abstract: Areal deformation monitoring based on point clouds can be a very valuable alternative to the established point-based monitoring techniques, especially for deformation monitoring of natural scenes. Other approaches incorporate some local geometric information by constraining the search for corresponding points along the direction of the normal vector of either the triangulated surface (C2M and M2M) or a plane fitted to the neighboring points (M3C2) These naïve ways of establishing the correspondences typically result in underestimation of the displacement magnitudes in parts of the point clouds that have changed (see Section 3). We show on a real case data set that the proposed method, which relies solely on the geometric information intrinsically available in point clouds, is able to correctly estimate 3D displacement vectors, while the traditional deformation models greatly underestimate the magnitudes and are incapable of determining the direction of the displacement vectors (Section 3.2.2). F2S3: Robustified determination of 3D displacement vector fields using deep learning | 179
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.