Dimensional Point Clouds Research Articles

Automated Inspection of Railway Tunnels’ Power Line Using LiDAR Point Clouds

Transport networks need periodic inspections to increase their safety and improve their management. In the last few years, LiDAR (light detection and ranging) technology has become a tool for helping to create a precise database of almost any type of infrastructure. Mobile laser scanning (MLS) systems use a laser beam to collect dense three dimensional (3D) point clouds, which include geometric and radiometric data of the environment in which they are placed. In the context of this paper, a methodology for automatically inspecting the clearance gauge and the deflection of the aerial contact line in railway tunnels is presented. The main objective is to compare results and verify their compliance with the Spanish norm. The 3D data are provided by a LYNX Mobile Mapper System (MMS). First, the area is surveyed and then the obtained (3D) point cloud is classified into contact wire, suspension wire, and remaining points. Finally, the inspection of the railway’s power line is performed. The validation of the proposed methodology has been carried out in three different tunnel point clouds, obtaining both qualitative and quantitative results for points’ classification, together with the results of the measures performed.

APPLICABILITY ASSESSMENT OF UAV MAPPING FOR DISASTER DAMAGE INVESTIGATION IN KOREA

Abstract. As natural disaster occurs, the local and the central government should investigate the damaged fields promptly, analyze quantitatively the degree of damage, and establish an appropriate disaster recovery plan in accordance with Framework Act on the Management of Disasters and Safety in Korea. The purpose of this study is to assess the applicability of UAV photogrammetry for the management of natural disaster. First, we suggest small easy-to-use UAV-based investigation procedure for natural disaster damaged area in the phase of disaster recovery in Korea. Before drone-based aerial surveying, the field survey can be performed with DGPS RTK for GCPs setting-up around disaster site. In this paper, we generate three dimensional terrain information and high-resolution ortho-imagery and then analyse quantitatively damage degree by natural disaster using commercial UAVs and drone mapping technique. Finally, we evaluate the mapping accuracy and work efficiency of drone mapping for disaster investigation application through comparing with traditional investigation work process which was dependent on labour-intensive field survey. The resolution ortho-image map of within less 5cm of GSD generated by aerial photographs acquired from UAVs at the altitude of 100m~250m enabled us to check damage information such as facilities destroy or the trace of soil erosion around the river flooded and reservoir collapsed area. In addition, three dimensional point cloud data of landslide-damaged areas enabled us to more accurately measure the width and the depth of outflows caused by landslides, soil runoff distance, and landslide damage area. The photogrammetry-based drone mapping technology for the disaster damage investigation is expected to be an alternative approach to support or replace the labour-intensive disaster site survey that needs to investigate the disaster site quickly and timely.

Three dimensional mapping of forest canopy equivalent water thickness using dual-wavelength terrestrial laser scanning

Globally, forests are being subjected to numerous threats, including climate change, wildfires, and insect and disease outbreaks, among others. Satellite optical remote sensing data have been widely utilized in early detection of tree and forest stress by estimating water status metrics such as the leaf Equivalent Water Thickness (EWT). This estimate, however, is affected by soil characteristics and understory vegetation and often ignores the effects of the fine-scale heterogeneity of canopy structure and leaf water content. Such effects can be better understood by studying the EWT distribution in three dimensions. In this study, Terrestrial Laser Scanning (TLS) intensity data from the commercially-available Leica P20 and P40 instruments (808 nm and 1550 nm respectively) were combined in a Normalized Difference Index (NDI). NDI was used to map EWT of 12 trees in three dimensions from floor to canopy in a mixed broadleaf forest plot (Wytham Woods, UK). The average error in EWT estimates across three species was less than 8%. The three dimensional point clouds revealed that, in this snapshot, EWT changes vertically, usually increasing towards canopy top. The proposed method has the potential to provide predawn EWT measurements, is independent of solar illumination, and can lead to a better understanding of the factors affecting satellite estimation of EWT.

SEMANTIC SEGMENTATION OF INDOOR 3D POINT CLOUD WITH SLENET

Abstract. With the rapid development of new indoor sensors and acquisition techniques, the amount of indoor three dimensional (3D) point cloud models was significantly increased. However, these massive “blind” point clouds are difficult to satisfy the demand of many location-based indoor applications and GIS analysis. The robust semantic segmentation of 3D point clouds remains a challenge. In this paper, a segmentation with layout estimation network (SLENet)-based 2D–3D semantic transfer method is proposed for robust segmentation of image-based indoor 3D point clouds. Firstly, a SLENet is devised to simultaneously achieve the semantic labels and indoor spatial layout estimation from 2D images. A pixel labeling pool is then constructed to incorporate the visual graphical model to realize the efficient 2D–3D semantic transfer for 3D point clouds, which avoids the time-consuming pixel-wise label transfer and the reprojection error. Finally, a 3D-contextual refinement, which explores the extra-image consistency with 3D constraints is developed to suppress the labeling contradiction caused by multi-superpixel aggregation. The experiments were conducted on an open dataset (NYUDv2 indoor dataset) and a local dataset. In comparison with the state-of-the-art methods in terms of 2D semantic segmentation, SLENet can both learn discriminative enough features for inter-class segmentation while preserving clear boundaries for intra-class segmentation. Based on the excellence of SLENet, the final 3D semantic segmentation tested on the point cloud created from the local image dataset can reach a total accuracy of 89.97%, with the object semantics and indoor structural information both expressed.

An Automated Rectification Method for Unmanned Aerial Vehicle LiDAR Point Cloud Data Based on Laser Intensity

Point cloud rectification is an efficient approach to improve the quality of laser point cloud data. Conventional rectification methods mostly relied on ground control points (GCPs), typical artificial ground objects, and raw measurements of the laser scanner which impede automation and adaptability in practice. This paper proposed an automated rectification method for the point cloud data that are acquired by an unmanned aerial vehicle LiDAR system based on laser intensity, with the goal to reduce the dependency of ancillary data and improve the automated level of the rectification process. First, laser intensity images were produced by interpolating the intensity data of all the LiDAR scanning strips. Second, a scale-invariant feature transform algorithm was conducted to extract two dimensional (2D) tie points from the intensity images; the pseudo tie points were removed by using a random sample consensus algorithm. Next, all the 2D tie points were transformed to three dimensional (3D) point cloud to derive 3D tie point sets. After that, the observation error equations were created with the condition of coplanar constraints. Finally, a nonlinear least square algorithm was applied to solve the boresight angular error parameters, which were subsequently used to correct the laser point cloud data. A case study in Shehezi, Xinjiang, China was implemented with our proposed method and the results indicate that our method is efficient to estimate the boresight angular error between the laser scanner and inertial measurement unit. After applying the results of the boresight angular error solution to rectify the laser point cloud, the planar root mean square error (RMSE) is 5.7 cm and decreased by 1.1 cm in average; the elevation RMSE is 1.4 cm and decreased by 0.8 cm in average. Comparing with the stepwise geometric method, our proposed method achieved similar horizontal accuracy and outperformed it in vertical accuracy of registration.

Tribocorrosion properties of polycrystalline diamond compact in saline environment

Tribocorrosion properties of polycrystalline diamond compact (PDC) in solution containing sodium chloride were investigated using a modified pin-on-disc wear test rig. Tests were conducted at various salinities (0–20%) for a better understanding of the abrasion-corrosion interaction. The corrosion behavior was examined through a three-electrode electrochemical cell. The contributions of pure abrasion and pure corrosion as well as their synergy to the total volume loss were calculated. The influence of cathodic protection on the abrasion-corrosion performance of PDC was also evaluated. Experimental data in this paper shows that the PDC suffers more severe damage as the sodium chloride concentration increases gradually and the applied potential of −1.0 V vs. SCE can suppress the corrosion-related effects to a negligible level. Both electrochemical measurements and SEM micrographs of worn surfaces were used to detail the degradation mechanism. Despite the fact that it is the mechanical wear that dominates the abrasion-corrosion mechanism, results reveal that the interaction between the abrasion and corrosion plays a significant role. Moreover, an effective volume calculation method based on three dimensional point cloud was proposed to determine the wear volume of PDC. Validation studies demonstrate that the measurement method is feasible and accurate with an error of no more than 5%.

Optimised MRI intensity standardisation based on multi-dimensional sub-regional point cloud registration

ABSTRACTAs one of the most widely used medical imaging methods, magnetic resonance imaging (MRI) and its corresponding automatic image analysis are research hotspots in computer-aided medical diagnosis. However, due to the disparities in imaging parameters and scanner characteristics, the variation in intensity distribution between scanners results in performance reduction in automatic image analysis and diagnosis. This paper aims at obtaining sub-region intensity distribution and forming a robust non-rigid intensity transforming function in order to standardise the intensities of MR images. The proposed method includes multi-modality image registration, sub-region standard intensity estimation, weighted -dimensional point cloud generation and global intensity transformation. Between the target images and reference images, this novel method could not only avoid the intensity distortion due to the inconsistency of inter-tissue brightness relationship, but also reduce the dependence on the accuracy of multi-modality MR image registration. Experiments show performance enhancement in peak signal-to-noise ratio, average disparity, histogram correlation, mean square error and structural similarity over the existing methods. Therefore, the intensities of MR images from different scanners could be standardised by the proposed method, so that multi-centre/multi-machine correlation could be promoted.

Three Dimensional Point Cloud Compression and Decompression Using Polynomials of Degree One

The availability of cheap depth range sensors has increased the use of an enormous amount of 3D information in hand-held and head-mounted devices. This has directed a large research community to optimize point cloud storage requirements by preserving the original structure of data with an acceptable attenuation rate. Point cloud compression algorithms were developed to occupy less storage space by focusing on features such as color, texture, and geometric information. In this work, we propose a novel lossy point cloud compression and decompression algorithm that optimizes storage space requirements by preserving geometric information of the scene. Segmentation is performed by using a region growing segmentation algorithm. The points under the boundary of the surfaces are discarded that can be recovered through the polynomial equations of degree one in the decompression phase. We have compared the proposed technique with existing techniques using publicly available datasets for indoor architectural scenes. The results show that the proposed novel technique outperformed all the techniques for compression rate and RMSE within an acceptable time scale.

Efficient speaker identification using spectral entropy

In voice recognition, the two main problems are speech recognition (what was said), and speaker recognition (who was speaking). The usual method for speaker recognition is to postulate a model where the speaker identity corresponds to the parameters of the model, which estimation could be time-consuming when the number of candidate speakers is large. In this paper, we model the speaker as a high dimensional point cloud of entropy-based features, extracted from the speech signal. The method allows indexing, and hence it can manage large databases. We experimentally assessed the quality of the identification with a publicly available database formed by extracting audio from a collection of YouTube videos of 1,000 different speakers. With 20 second audio excerpts, we were able to identify a speaker with 97% accuracy when the recording environment is not controlled, and with 99% accuracy for controlled recording environments.

Evaluation and characterization of deterministic laser textured surfaces using machine vision

Texturing of surfaces is often done to improve the surface properties. The purpose of this study is to develop a methodology for two dimensional inspection of the quality of textured surfaces automatically using machine vision techniques. The quality is measured in an absolute sense, using the value of offset between ideal shapes that inscribe and circumscribe the boundary of a machined feature. This gives an idea about machining accuracy rather than similarity between the machined texture shape and intended ideal shape which most of the existing methods try to measure. The techniques developed are tested on textures created on polished titanium surface using laser micro machining technique. In this study the textured surfaces made up of four different shapes namely circle, ellipse, triangle and square are considered. The automatic classification of texture shapes into the mentioned categories is achieved by finding their correlation with ideal shapes. For the purpose of measuring the machining accuracy, this paper extends the ISO definition of circularity error to ellipse, triangle and square shapes and proposes a method to fit least square triangle and square shapes to a given two dimensional point clouds. The error and the average dimension of the features are calculated from circumscribing, inscribing and least square fit shapes. The results generated by the proposed algorithms are found to be in good agreement with the known dimension values set during machining of the samples and manual measurements done using a stereo-microscope images. This indicate the suitability of this procedure for characterization of textured surfaces for different applications.

EVALUATING THE PERFORMANCE OF TERRESTRIAL LASER SCANNING FOR LANDSLIDE MONITORING

Abstract. Nowadays, Terrestrial Laser Scanning (TLS) technology is gaining popularity in monitoring and predicting the movement of landslide due to the capability of high-speed data capture without requiring a direct contact with the monitored surface. It offers very high density of point cloud data in high resolution and also can be an effective tool in detecting the surface movement of the landslide area. The aim of this research is to determine the optimal level of scanning resolution for landslide monitoring using TLS. The Topcon Geodetic Laser Scanner (GLS) 2000 was used in this research to obtain the three dimensional (3D) point cloud data of the landslide area. Four types of resolution were used during scanning operation which were consist of very high, high, medium and low resolutions. After done with the data collection, the point clouds datasets were undergone the process of registration and filtering using ScanMaster software. After that, the registered point clouds datasets were analyzed using CloudCompare software. Based on the results obtained, the accuracy of TLS point cloud data between picking point manually and computed automatically by ScanMaster software shows the maximum Root Mean Square (RMS) value of coordinate differences were 0.013 m in very high resolution, 0.017 m in high resolution, 0.031 m in medium resolution and 0.052 m in low resolution respectively. Meanwhile, the accuracy of TLS point cloud data between picking point manually and total station data using intersection method shows the maximum RMS values of coordinate differences were 0.013 m in very high resolution, 0.018 m in high resolution, 0.033 m in medium resolution and 0.054 m in low resolution respectively. Hence, it can be concluded that the high or very high resolution is needed for landslide monitoring using Topcon GLS-2000 which can provide more accurate data in slope result, while the low and medium resolutions is not suitable for landslide monitoring due to the accuracy of TLS point cloud data that will decreased when the resolution value is increased.

Analyzing the Squared Distance-to-Measure Gradient Flow System with k-Order Voronoi Diagrams

Point cloud data arises naturally from 3-dimensional scanners, LiDAR sensors, and industrial computed tomography (i.e. CT scans) among other sources. Most point clouds obtained through experimental means exhibit some level of noise, inhibiting mesh reconstruction algorithms and topological data analysis techniques. To alleviate the problems caused by noise, smoothing algorithms are often employed as a preprocessing step. Moving least squares is one such technique, however, many of these techniques are designed to work on surfaces in $$\mathbb {R}^3$$ . As interesting point clouds can naturally live in higher dimensions, we seek a method for smoothing higher dimensional point clouds. To this end, we turn to the distance to measure function. In this paper, we provide a theoretical foundation for studying the gradient flow induced by the squared distance to measure function, as introduced by Chazal, Cohen-Steiner, and Merigot. In particular, we frame the gradient flow as a Filippov system and find a method for solving the squared distance to measure gradient flow, induced by the uniform empirical measure, using higher order Voronoi diagrams. In contrast to some existing techniques, this gradient flow provides a smoothing algorithm which computationally scales with dimensionality.

MANAGEMENT OF LARGE INDOOR POINT CLOUDS: AN INITIAL EXPLORATION

Abstract. Indoor navigation and visualization become increasingly important nowadays. Meanwhile, the proliferation of new sensors as well as the advancement of data processing provide massive point clouds to model the indoor environment in high accuracy. However, current state-of-the-art solutions fail to manage such large datasets efficiently. File based solutions often require substantial development work while database solutions are still faced with issues such as inefficient data loading and indexing. In this research, through a case study which aims to solve the problem of intermittent rendering of massive points in the context of indoor navigation, we devised and implemented an algorithm to compute the continuous Level of Detail (cLoD) where geometric and classification information are considered. Benchmarks are developed and different approaches in Oracle are tested to learn the pros and cons. Surprisingly, the flat table approach could be very efficient compared with other schemes. The crucial point lies in how to address priority of different dimensions including cLoD, classification and spatial dimensions, and avoid unnecessary scanning of the table. Writing results either to the memory or the disk constitutes major part of the time cost when large output is concerned. Conventional solutions based on spatial data objects present poor performance due to cumbersome indexing structure, inaccurate selection and additional decoding process. Besides, approximate selection in the unit of physical object is proposed and the performance is satisfactory when large amount of data is requested. The knowledge acquired could prompt the development of a novel data management of high dimensional point clouds where the classification information is involved.

Toward better boundary preserved supervoxel segmentation for 3D point clouds

Supervoxels provide a more natural and compact representation of three dimensional point clouds, and enable the operations to be performed on regions rather than on the scattered points. Many state-of-the-art supervoxel segmentation methods adopt fixed resolution for each supervoxel, and rely on the initialization of seed points. As a result, they may not preserve well the boundaries of the point cloud with a non-uniform density. In this paper, we present a simple but effective supervoxel segmentation method for point clouds, which formalizes supervoxel segmentation as a subset selection problem. We develop an heuristic algorithm that utilizes local information to efficiently solve the subset selection problem. The proposed method can produce supervoxels with adaptive resolutions, and dose not rely the selection of seed points. The method is fully tested on three publicly available point cloud segmentation benchmarks, which cover the major point cloud types. The experimental results show that compared with the state-of-the-art supervoxel segmentation methods, the supervoxels extracted using our method preserve the object boundaries and small structures more effectively, which is reflected in a higher boundary recall and lower under-segmentation error.

Iterative individual plant clustering in maize with assembled 2D LiDAR data

A two dimensional (2D) laser scanner was mounted at the front part of a small 4-wheel autonomous robot with differential steering, at an angle of 30 ° pointing downwards. The machine was able to drive between maize rows and collect concurrent time-stamped data. A robotic total station tracked the position of a prism mounted on the vehicle. The total station and laser scanner data were fused to generate a three dimensional (3D) point cloud. This 3D representation was used to detect individual plant positions, which are of particular interest for applications such as phenotyping, individual plant treatment and precision weeding. Two different methodologies were applied to the 3D point cloud to estimate the position of the individual plants. The first methodology used the Euclidian Clustering on the entire point cloud. The second methodology utilised the position of an initial plant and the fixed plant spacing to search iteratively for the best clusters. The two algorithms were applied at three different plant growth stages. For the first method, results indicated a detection rate up to 73.7% with a root mean square error of 3.6 cm. The second method was able to detect all plants (100% detection rate) with an accuracy of 2.7–3.0 cm, taking the plant spacing of 13 cm into account.

Research on Tomato Nitrogen Content Nondestructive Testing Method Based on Multidimensional Image Processing Technology

This paper is aimed at greenhouse tomato nitrogen detection using hyperspectral imaging combined with three dimensional laser scanning technology. This technology extracts the nitrogen hyperspectral feature image and the plant three dimensional morphological characters, to achieve the rapid quantitative analysis of nitrogen in tomato. The characteristic spectrum of nitrogen was extracted, and the mean intensity characteristic of the image feature was obtained. Then based on the acquisition of the tomato hyperspectral image data cube at different nitrogen levels, the sensitive region stepwise regression combined with correlation analysis was performed. Based on the acquired three dimensional laser scanning data of tomatoes, the stem diameter, the plant height and other biomass characteristics of different nitrogen levels were obtained by establishing the spatial geometric model of tomato three dimensional point cloud. A multi-feature fusion model for tomato nitrogen detection was established by partial least square regression. The results showed that the R2 in the constructed model was 0.94, with the accuracy significantly better than that of the single feature model established by using hyperspectral image and three dimensional laser scanning.

Outliers Detection Method Based on Dynamic Standard Deviation Threshold Using Neighborhood Density Constraints for Three Dimensional Point Cloud

Outliers Detection Method Based on Dynamic Standard Deviation Threshold Using Neighborhood Density Constraints for Three Dimensional Point Cloud

LIDAR-based fracture characterization and controlling factors analysis: An outcrop case from Kuqa Depression, NW China

Fracture is of primary importance to the natural gas production capacity from many tight sandstone reservoirs in the Kuqa foreland basin, NW China, but the orientation, size and plane porosity of the fractures in the subsurface is difficult to measure directly. Terrestrial light detection and ranging (LIDAR) surveys can offer factual information of outcrop-based research efforts to characterize fracture development laws and controlling factors. In this paper, utilizing a multi-level covering, we obtain the three dimensional (3D) point cloud data from a LIDAR survey launched at a typical outcrop. Matching with high-resolution digital photos and artificial measured information, the 3-D positions of natural fractures are extracted strictly in the data volume section. Furthermore, the fracture and reservoir model can be founded based on the systematic sampling and laboratory analysis, while a variety of accurate facture parameters can be obtained. It is founded that three groups of shear fractures are mainly developed in two periods with large inclination, short trace length and small spacing of normal distribution. Its patterns has provided a literal distribution of penetrating fracture zone and interlayer fracture zone with a single and a double set of advantage orientation respectively. It turned out that the fracture development scale is controlled by lithology, layer thickness, maximum principle paleostress and rock composition with a good exponential relationship. Our work could provide a workflow linking outcrop fracture observations to the 3D model of subsurface fracture prediction and extend modeling capability in other outcrop studies.

MODEL FOR SEMANTICALLY RICH POINT CLOUD DATA

Abstract. This paper proposes an interoperable model for managing high dimensional point clouds while integrating semantics. Point clouds from sensors are a direct source of information physically describing a 3D state of the recorded environment. As such, they are an exhaustive representation of the real world at every scale: 3D reality-based spatial data. Their generation is increasingly fast but processing routines and data models lack of knowledge to reason from information extraction rather than interpretation. The enhanced smart point cloud developed model allows to bring intelligence to point clouds via 3 connected meta-models while linking available knowledge and classification procedures that permits semantic injection. Interoperability drives the model adaptation to potentially many applications through specialized domain ontologies. A first prototype is implemented in Python and PostgreSQL database and allows to combine semantic and spatial concepts for basic hybrid queries on different point clouds.

Not All Trees Sleep the Same-High Temporal Resolution Terrestrial Laser Scanning Shows Differences in Nocturnal Plant Movement.

Circadian leaf movements are widely known in plants, but nocturnal movement of tree branches were only recently discovered by using terrestrial laser scanning (TLS), a high resolution three-dimensional surveying technique. TLS uses a pulsed laser emitted in a regular scan pattern for rapid measurement of distances to the targets, thus producing three dimensional point cloud models of sub-centimeter resolution and accuracy in a few minutes. Here, we aim to gain an overview of the variability of circadian movement of small trees across different taxonomic groups, growth forms and leaf anatomies. We surveyed a series of 18 full scans over a 12-h night period to measure nocturnal changes in shape simultaneously for an experimental setup of 22 plants representing different species. Resulting point clouds were evaluated by comparing changes in height percentiles of laser scanning points belonging to the canopy. Changes in crown shape were observed for all studied trees, but clearly distinguishable sleep movements are apparently rare. Ambient light conditions were continuously dark between sunset (7:30 p.m.) and sunrise (6:00 a.m.), but most changes in movement direction occurred during this period, thus most of the recorded changes in crown shape were probably not controlled by ambient light. The highest movement amplitudes, for periodic circadian movement around 2 cm were observed for Aesculus and Acer, compared to non-periodic continuous change in shape of 5 cm for Gleditschia and 2 cm for Fargesia. In several species we detected 2–4 h cycles of minor crown movement of 0.5–1 cm, which is close to the limit of our measurement accuracy. We present a conceptual framework for interpreting observed changes as a combination of circadian rhythm with a period close to 12 h, short-term oscillation repeated every 2–4 h, aperiodic continuous movement in one direction and measurement noise which we assume to be random. Observed movement patterns are interpreted within this framework, and connections with morphology and taxonomy are proposed. We confirm the existence of overnight “sleep” movement for some trees, but conclude that circadian movement is a variable phenomenon in plants, probably controlled by a complex combination of anatomical, physiological, and morphological factors.