Silverio García-Cortés

Automatic Detection and Classification of Pole-Like Objects in Urban Point Cloud Data Using an Anomaly Detection Algorithm

Detecting and modeling urban furniture are of particular interest for urban management and the development of autonomous driving systems. This paper presents a novel method for detecting and classifying vertical urban objects and trees from unstructured three-dimensional mobile laser scanner (MLS) or terrestrial laser scanner (TLS) point cloud data. The method includes an automatic initial segmentation to remove the parts of the original cloud that are not of interest for detecting vertical objects, by means of a geometric index based on features of the point cloud. Vertical object detection is carried out through the Reed and Xiaoli (RX) anomaly detection algorithm applied to a pillar structure in which the point cloud was previously organized. A clustering algorithm is then used to classify the detected vertical elements as man-made poles or trees. The effectiveness of the proposed method was tested in two point clouds from heterogeneous street scenarios and measured by two different sensors. The results for the two test sites achieved detection rates higher than 96%; the classification accuracy was around 95%, and the completion quality of both procedures was 90%. Non-detected poles come from occlusions in the point cloud and low-height traffic signs; most misclassifications occurred in man-made poles adjacent to trees.

Morphological Operations to Extract Urban Curbs in 3D MLS Point Clouds

Automatic curb detection is an important issue in road maintenance, three-dimensional (3D) urban modeling, and autonomous navigation fields. This paper is focused on the segmentation of curbs and street boundaries using a 3D point cloud captured by a mobile laser scanner (MLS) system. Our method provides a solution based on the projection of the measured point cloud on the XY plane. Over that plane, a segmentation algorithm is carried out based on morphological operations to determine the location of street boundaries. In addition, a solution to extract curb edges based on the roughness of the point cloud is proposed. The proposed method is valid in both straight and curved road sections and applicable both to laser scanner and stereo vision 3D data due to the independence of its scanning geometry. The proposed method has been successfully tested with two datasets measured by different sensors. The first dataset corresponds to a point cloud measured by a TOPCON sensor in the Spanish town of Cudillero. The second dataset corresponds to a point cloud measured by a RIEGL sensor in the Austrian town of Horn. The extraction method provides completeness and correctness rates above 90% and quality values higher than 85% in both studied datasets.

Study of a rockfall in a limy conglomerate canyon (Covarrubias, Burgos, N. Spain)

In this paper, a rockfall of some tenths of cubic metres of rock, occurred in a conglomerate canyon in the northern Castilian plateau (N-Central Spain), is described in detail. This includes the basic data (geology, mapping, meteorological data and rock involved), the mechanical mechanism of the instability, the propagation and reach of the fallen rock boulders and the final location of the pieces. The study is based on detailed geological, aerial photography and geotechnical studies and advance topographical measurements. Geological engineering techniques are used to analyze the instability initiation and rockfall propagation. This study contributes to a better understanding of the geological processes leading to the occurrence of calcareous canyons and cliffs in the drainage network of the upper Castilian plateau. It is also helpful to improve our knowledge on natural rockfall phenomena for risk engineering purposes aiming to quantify, control and mitigate associated hazards.

A statistical method for geometry inspection from point clouds

Abstract This paper introduces a statistical methodology for geometry inspection from point clouds obtained with a laser scanner or other measuring systems. The null hypothesis of interest is that the real surface of an object fits the theoretical shape and dimensions of the object. An algorithm based on bivariate kernel smoothers is used to nonparametrically estimate the surface of the object and bootstrap-based procedures are proposed for testing the null hypothesis. In order to validate the methodology a simulated study was conducted. Finally, the proposed methodology was applied to the inspection of a parabolic dish antenna.

Detection and magnification of bridge displacements using video images

Monitoring displacements on some structures such as large bridges is essential to study their structural performance in order to avoid severe damage or even their collapse. In this work, we use images obtained with digital video cameras to estimate the displacements of a metallic bridge by means of cross-correlation. Thus, it was possible to detect millimetric displacements for distances between the camera and the bridge upper ten meters. In order to obtain a better representation of the structural displacements along the bridge and its modal shapes, a technique of video magnification was also applied. The results obtained show that the combination of both techniques can provide relevant information for a structural analysis of the bridge.

Stability analysis of a tunnel using LIDAR data and the keyblock method

We describe a stability study for a tunnel excavated in a fractured rock mass. The proposed methodology uses automatic and semi-automatic algorithms to extract joint sets from a LIDAR point cloud. Using that information and considering a model of ubiquitous joints, pyramidal and non-pyramidal unstable blocks were identified using the keyblock method. The algorithms used for the stability analysis were implemented in BLOFOR software, which represented the possible location of the blocks in the tunnel in three dimensions and provided information on tunnel stability, including a safety factor for each block.

A novel geometric approach for 3-D geological modelling

This paper presents a graphics system called the prism network and its boundary (PNB) aimed at representing the stratified solids that are formed in the subsurface terrain. It allows full 3D modelling of these types of spaces formed by superimposed layers bounded by stratification surfaces. A set of boreholes in the ground is used to generate a network of prisms that models the solid space and serves as the basis for creating two secondary networks, one of triangles and another of quadrilaterals, which define the model boundaries. This graphics system has been implemented in C++, using the standard OpenGL graphics API. The paper first defines and implements a model in the PNB data structure, the process by means of which it is generated is then described, and, finally, a range of applications to specific cases is presented and graphically illustrated.RésuméCet article présente un système graphique appelé PNB (sigles en anglais de Réseau de Prismes et sa Frontière) qui vise à représenter les solides stratifiés qui sont formés dans le sous-sol. Il permet la modélisation 3D complète de ces types d’espaces formés par des couches superposées délimitées par des surfaces de stratification. À partir d’un ensemble de forages dans le terrain est généré un réseau de prismes qui modélise l’espace solide et sert de base pour la création de deux réseaux secondaires, l’un de triangles et l’autre de quadrilatères, qui définissent les frontières du modèle. Ce système graphique a été implémenté en C++, en utilisant l’API standard graphique OpenGL. Le article définit d’abord le modèle et le met en œuvre dans la structure de données PNB, puis décrit le processus de leur génération, et enfin, présente et illustre graphiquement une gamme d’applications à des cas particuliers.

A mathematical algorithm for dimensional control of tunnels using topographic profiles

We propose a new method to study the geometry of tunnels based on comparing topographic profiles considered as curves instead of sets of discrete points. Two different approaches were used to compare the profiles: one based on the notion of functional depth, which is a measure of the centrality of a set of curves, and the other based on ordering the scores of principal component decomposition. The proposed methodology was used to analyse the geometric quality of a tunnel by comparing a set of cross-sections of the tunnel measured using a RIEGL LMS-Z390i terrestrial laser scanner. The results obtained demonstrate the usefulness of the method based on principal component decomposition in detecting significant differences of the geometry along the tunnel.

Detecting imperceptible movements in structures by means of video magnification

The naked eye is not able to perceive very slow movements such as those occurring in certain structures under external forces. This might be the case of metallic or concrete bridges, tower cranes or steel beams. However, sometimes it is of interest to view such movements, since they can provide useful information regarding the mechanical state of those structures. In this work, we analyze the utility of video magnification to detect imperceptible movements in several types of structures. First, laboratory experiments were conducted to validate the method. Then, two different tests were carried out on real structures: one on a water slide and another on a tower crane. The results obtained allow us to conclude that image cross-correlation and video magnification is indeed a promising low-cost technique for structure health monitoring.

Automatic road edge detection from Mobile Laser Scanning (MLS)

In this article we present an algorithm for automatic road edge detection from MLS (Mobile Laser Scanning) data. The method takes advantage of linear structures derived from MLS point clouds. These lines are extracted from the point cloud and grouped following geometric restrictions. Then, the outlines of the groups are extracted as road edges. Finally, a moving window filter is applied to those points in order to remove outliers and delineate the road edge. The method was tested on an 800m stretch of road, and the results were checked through visual inspection. Correctness and completeness were 99.1% and 97.5%, respectively.