Classification of Terrestrial Laser Scanner Point Clouds: A Comparison of Methods for Landslide Monitoring from Mathematical Surface Approximation

Abstract

Terrestrial laser scanners (TLS) are contact-free measuring sensors that record dense point clouds of objects or scenes by acquiring coordinates and an intensity value for each point. The point clouds are scattered and noisy. Performing a mathematical surface approximation instead of working directly on the point cloud is an efficient way to reduce the data storage and structure the point clouds by transforming “data” to “information”. Applications include rigorous statistical testing for deformation analysis within the context of landslide monitoring. In order to reach an optimal approximation, classification and segmentation algorithms can identify and remove inhomogeneous structures, such as trees or bushes, to obtain a smooth and accurate mathematical surface of the ground. In this contribution, we compare methods to perform the classification of TLS point clouds with the aim of guiding the reader through the existing algorithms. Besides the traditional point cloud filtering methods, we will analyze machine learning classification algorithms based on the manual extraction of point cloud features, and a deep learning approach with automatic extraction of features called PointNet++. We have intentionally chosen strategies easy to implement and understand so that our results are reproducible for similar point clouds. We show that each method has advantages and drawbacks, depending on user criteria, such as the computational time, the classification accuracy needed, whether manual extraction is performed or not, and if prior information is required. We highlight that filtering methods are advantageous for the application at hand and perform a mathematical surface approximation as an illustration. Accordingly, we have chosen locally refined B-splines, which were shown to provide an optimal and computationally manageable approximation of TLS point clouds.

Monitoring and Quantifying Soil Erosion and Sedimentation Rates in Centimeter Accuracy Using UAV-Photogrammetry, GNSS, and t-LiDAR in a Post-Fire Setting

“…Current research denotes the utilization of combined multitemporal UAV images and Structure from Motion (SfM) technologies to systematically monitor both the topsoil surface and volume change within small-scale watersheds [56,124]. In particular, recent studies [125][126][127] focus on point cloud analysis. As [108] indicated, the DEM cannot accurately represent the complex surface due to the 2D data used, while other researchers [128] highlight the efficiency of point cloud analysis in volumetric calculations.…”

A review on how Big Data can help to monitor the environment and to mitigate risks due to climate change

“…In a related context, [30] applied a change point detection method to identify seeds for point cloud segmentation. Besides the conventional point cloud filtering techniques, [31] and [32] explored various ML classification algorithms, such as the deep learning approach incorporating automatic feature extraction (PointNet++). In [33], the advantages of an unsupervised ML approach called Gaussian Mixture Modeling, were demonstrated.…”

“…In the near future, automatic deformation analysis and timely, near-realtime risk assessments will be conducted using the dedicated ML and enhanced segmentation approaches mentioned earlier, which are currently under development. Further, new methods based on volume approximation for data reduction and visualization open new possibilities, facilitating the implementation of ML techniques and the transformation of the Big Data to information [31]. Similarly, recent studies have developed innovative approaches to monitor and forecast geohazards combining ML algorithms and various datasets.…”

Comparative Analysis of Algorithms to Cleanse Soil Micro-Relief Point Clouds

“…Filtering algorithms simply extract subsets of point clouds based on specific criteria like intensity or reflectance values. In contrast to the segmentation process, not all filtering algorithms need neighbourhood as feature information [14]. In contrast, classification methods use statistical or machine learning methods [14].…”

“…In contrast to the segmentation process, not all filtering algorithms need neighbourhood as feature information [14]. In contrast, classification methods use statistical or machine learning methods [14].…”