3D LiDAR Point Cloud Research Articles

Correcting NLOS by 3D LiDAR and building height to improve GNSS single point positioning

We present a novel method to detect the GNSS NLOS and correct the NLOS pseudorange measurements based on on-board sensing. This paper demonstrates the use of LiDAR scanner and a list of building heights to describe the perceived environment. To estimate the geometry and pose of the top edges of buildings (TEBs) relative to the GNSS receiver, a surface segmentation method is employed to detect the TEBs of surrounding buildings using 3D LiDAR point clouds. The top edges of the building are extracted and extended using the building height list in Skyplot to identify the NLOS-affected ones. Innovatively, the NLOS delay in pseudorange is corrected based on the detected TEBs. Weighted least squares (WLS) is used to cooperate the corrected NLOS and other pseudorange measurements. Vehicle experiments are conducted in two different urban canyons to verify the effectiveness of the proposed method in improving GNSS single point positioning (SPP) accuracy.

Integrating Dense LiDAR-Camera Road Detection Maps by a Multi-Modal CRF Model

Road detection is an important task in autonomous navigation systems. In this paper, we propose a road detection method via a LiDAR-camera fusion strategy to exploit both the range and color information. The whole system consists of three parts. In the LiDAR based part, we transform the discrete 3D LiDAR point clouds to continuous 2D LiDAR range images and propose a distance-aware height-difference based scanning approach to get the road estimations quickly. In the camera based part, we apply a light-weight transfer learning based road segmentation network. In the LiDAR-camera fusion part, we transform the detection results from LiDAR and camera to dense and binary ones to solve the data imbalance problem and fuse them in a multi-modal conditional random field (MM-CRF) framework. Experiments show that the proposed MM-CRF fusion method can operate in real-time and achieve competitive performance compared with the state-of-the-art road detection algorithms on the KITTI-Road benchmark.

Deep learning for conifer/deciduous classification of airborne LiDAR 3D point clouds representing individual trees

The purpose of this study was to investigate the use of deep learning for coniferous/deciduous classification of individual trees segmented from airborne LiDAR data. To enable processing by a deep convolutional neural network (CNN), we designed two discrete representations using leaf-off and leaf-on LiDAR data: a digital surface model with four channels (DSM × 4) and a set of four 2D views (4 × 2D). A training dataset of tree crowns was generated via segmentation of tree crowns, followed by co-registration with field data. Potential mislabels due to GPS error or tree leaning were corrected using a statistical ensemble filtering procedure. Because the training data was heavily unbalanced (~8% conifers), we trained an ensemble of CNNs on random balanced sub-samples. Benchmarked against multiple traditional shallow learning methods using manually designed features, the CNNs improved accuracies up to 14%. The 4 × 2D representation yielded similar classification accuracies to the DSM × 4 representation (~82% coniferous and ~90% deciduous) while converging faster. Further experimentation showed that early/late fusion of the channels in the representations did not affect the accuracies in a significant way. The data augmentation that was used for the CNN training improved the classification accuracies, but more real training instances (especially coniferous) likely results in much stronger improvements. Leaf-off LiDAR data were the primary source of useful information, which is likely due to the perennial nature of coniferous foliage. LiDAR intensity values also proved to be useful, but normalization yielded no significant improvement. As we observed, large training data may compensate for the lack of a subset of important domain data. Lastly, the classification accuracies of overstory trees (~90%) were more balanced than those of understory trees (~90% deciduous and ~65% coniferous), which is likely due to the incomplete capture of understory tree crowns via airborne LiDAR. In domains like remote sensing and biomedical imaging, where the data contain a large amount of information and are not friendly to human visual system, human-designed features may become suboptimal. As exemplified by this study, automatic, objective derivation of optimal features via deep learning can improve prediction tasks in such domains.

Automatic Tunnel Steel Arches Extraction Algorithm Based on 3D LiDAR Point Cloud.

Automation is an inevitable trend in the development of tunnel shotcrete machinery. Tunnel environmental perception based on 3D LiDAR point cloud has become a research hotspot. Current researches about the detection of tunnel point clouds focus on the completed tunnel with a smooth surface. However, few people have researched the automatic detection method for steel arches installed on a complex rock surface. This paper presents a novel algorithm to extract tunnel steel arches. Firstly, we propose a refined function for calibrating the tunnel axis by minimizing the density variance of the projected point cloud. Secondly, we segment the rock surface from the tunnel point cloud by using the region-growing method with the parameters obtained by analyzing the tunnel section sequence. Finally, a Directed Edge Growing (DEG) method is proposed to detect steel arches on the rock surface in the tunnel. Our experiment in the highway tunnels under construction in Changsha (China) shows that the proposed algorithm can effectively extract the points of the edge of steel arches from 3D LiDAR point cloud of the tunnel without manual assistance. The results demonstrated that the proposed algorithm achieved 92.1% of precision, 89.1% of recall, and 90.6% of the F-score.

Delaunay Triangulation-Based Spatial Clustering Technique for Enhanced Adjacent Boundary Detection and Segmentation of LiDAR 3D Point Clouds.

In spatial data with complexity, different clusters can be very contiguous, and the density of each cluster can be arbitrary and uneven. In addition, background noise that does not belong to any clusters in the data, or chain noise that connects multiple clusters may be included. This makes it difficult to separate clusters in contact with adjacent clusters, so a new approach is required to solve the nonlinear shape, irregular density, and touching problems of adjacent clusters that are common in complex spatial data clustering, as well as to improve robustness against various types of noise in spatial clusters. Accordingly, we proposed an efficient graph-based spatial clustering technique that employs Delaunay triangulation and the mechanism of DBSCAN (density-based spatial clustering of applications with noise). In the performance evaluation using simulated synthetic data as well as real 3D point clouds, the proposed method maintained better clustering and separability of neighboring clusters compared to other clustering techniques, and is expected to be of practical use in the field of spatial data mining.

PRECISE RADIONUCLIDE LOCALIZATION USING UAV-BASED LIDAR AND GAMMA PROBE WITH REAL-TIME PROCESSING

Abstract. In this work we propose an effective radiation source localization device employing a RIEGL VUX-1UAV laser scanner and a highly sensitive Hotzone Technologies gamma radiation probe mounted on a RiCOPTER UAV combined with real-time data processing. The on-board processing and radio communication system integrated within the UAV enables instant and continuously updated access to georeferenced 3D lidar point clouds and gamma radiation intensities. Further processing is done fully automated on the ground. We present a novel combination of both the 3D laser data and the gamma readings within an optimization algorithm that can locate the radioactive source in real-time. Furthermore, this technique can be used to estimate an on-ground radiation intensity, which also considers the actual topography as well as radiation barriers like vegetation or buildings. Results from field tests with real radioactive sources show that single sources can be located precisely, even if the source was largely covered. Outcomes are displayed to the person in charge in an intuitive and user-friendly way, e.g. on a tablet. The whole system is designed to operate in real-time and while the UAV is in the air, resulting in a highly flexible and possibly life-saving asset for firstresponders in time-critical scenarios.

Ground segmentation from 3D point cloud using features of scanning line segments

Aiming at the problem of accurately segmenting the ground in real-time from 3D LiDAR point cloud, a ground segmentation algorithm based on the features of scanning line segments is proposed. The algorithm first performs de-noising and pose correction on the 3D point cloud, then divides the scanning line according to the Euclidean distance and absolute height difference between adjacent points. Next, the characteristics of the adjacent line segments such as spacing, slope, and absolute height difference are analyzed. The maximum likelihood estimation is used to solve the feature threshold function, which improves the adaptability of threshold. Finally, comprehensively considering the undulating and inclined complex terrain, the scanning line segments are marked as segments of flat ground, segments of slope and segments of obstacle by formulating the new horizontal and vertical classification strategies. This algorithm has been successfully applied to the unmanned ground platform. The usage and comparative test show that the algorithm can detect the ground stably and efficiently in both urban and field scenarios.

Automated Visual Recognizability Evaluation of Traffic Sign Based on 3D LiDAR Point Clouds

Maintaining the high visual recognizability of traffic signs for traffic safety is a key matter for road network management. Mobile Laser Scanning (MLS) systems provide efficient way of 3D measurement over large-scale traffic environment. This paper presents a quantitative visual recognizability evaluation method for traffic signs in large-scale traffic environment based on traffic recognition theory and MLS 3D point clouds. We first propose the Visibility Evaluation Model (VEM) to quantitatively describe the visibility of traffic sign from any given viewpoint, then we proposed the concept of visual recognizability field and Traffic Sign Visual Recognizability Evaluation Model (TSVREM) to measure the visual recognizability of a traffic sign. Finally, we present an automatic TSVREM calculation algorithm for MLS 3D point clouds. Experimental results on real MLS 3D point clouds show that the proposed method is feasible and efficient.

SMART GREEN INFRASTRUCTURE IN A SMART CITY – THE CASE STUDY OF ECOSYSTEM SERVICES EVALUATION IN KRAKOW BASED ON I-TREE ECO SOFTWARE

It is a common perception that urban greenery does not bring any rational benefits, while profits from real estates are obvious. Therefore, the cities green infrastructure (urban forests, parks, trees, lawns, meadows, etc.) are constantly threatened with housing and development. However, urban greenery plays a substantial role in improving the quality of urbanites’ life, which is particularly significant in terms of predicted 70% urbanization rate by 2050. Healthy and well-managed city green infrastructure can improve air quality, remove particulate matters (PM) and CO2 sequestrate carbon, cool down the temperature or protect against winds. These functions of vegetation are known as ecosystem services (ES). Recognizing the value of ES provided by green infrastructure is crucial for urban planning and management in terms of assuring sustainable urban development. In our study, we used the i-Tree Eco (USDA Forest Service) software, which quantifies vegetation structure, environmental effects, and values of ES. The i-Tree Eco model is based on air pollution and local meteorological data along with the field data from the inventory of city vegetation. Requiring easy to collect (e.g. based on LiDAR 3D point clouds) input data and having a user-friendly interface, the i-Tree Eco has a potential of becoming a very useful tool for planners and managers in their everyday work. In this paper, we present a case study of ES evaluation for the “Krakowski Park” in Krakow (582 trees on 4.77 hectares, with the domination of Fraxinus excelsior, Ulmus laevis, and Betula pendula). For the analyzed 2015 year, the Krakowski Park trees stored in total 441.59 t of carbon, removed 184 kg of air pollutants and contributed to 220 m3 of avoided runoff. The total value of ecosystem services provided by Krakowski Park in the year 2015 was 5.096 EUR (8.76 EUR tree/year). In our further work, we intend to expand the ES evaluation on other green areas in Krakow and on a wider range of ES. Keywords: urban greenery, urban management and planning, air pollution, carbon sequestration, GIS http://doi.org/10.35180/gse-2019-0010

Field-Based High-Throughput Phenotyping for Maize Plant Using 3D LiDAR Point Cloud Generated With a "Phenomobile".

With the rapid rising of global population, the demand for improving breeding techniques to greatly increase the worldwide crop production has become more and more urgent. Most researchers believe that the key to new breeding techniques lies in genetic improvement of crops, which leads to a large quantity of phenotyping spots. Unfortunately, current phenotyping solutions are not powerful enough to handle so many spots with satisfying speed and accuracy. As a result, high-throughput phenotyping is drawing more and more attention. In this paper, we propose a new field-based sensing solution to high-throughput phenotyping. We mount a LiDAR (Velodyne HDL64-S3) on a mobile robot, making the robot a “phenomobile.” We develop software for data collection and analysis under Robotic Operating System using open source components and algorithm libraries. Different from conducting phenotyping observations with an in-row and one-by-one manner, our new solution allows the robot to move around the parcel to collect data. Thus, the 3D and 360° view laser scanner can collect phenotyping data for a large plant group at the same time, instead of one by one. Furthermore, no touching interference from the robot would be imposed onto the crops. We conduct experiments for maize plant on two parcels. We implement point cloud merging with landmarks and Iterative Closest Points to cut down the time consumption. We then recognize and compute the morphological phenotyping parameters (row spacing and plant height) of maize plant using depth-band histograms and horizontal point density. We analyze the cloud registration and merging performances, the row spacing detection accuracy, and the single plant height computation accuracy. Experimental results verify the feasibility of the proposed solution.

Mapping Riparian Vegetation Functions Using 3D Bispectral LiDAR Data

Riparian zones experience many anthropic pressures and are the subject of European legislation to encourage their monitoring and management, to attenuate these pressures. Assessing the effectiveness of management practices requires producing indicators of ecological functions. Laser Detection and Ranging (LiDAR) data can provide valuable information to assess the ecological status of riparian zones. The objective of this study was to evaluate the potential of LiDAR point clouds to produce indicators of riparian zone status. We used 3D bispectral LiDAR data to produce several indicators of a riparian zone of a dammed river in Normandy (France). The indicators were produced either directly from the 3D point clouds (e.g., biomass overhanging the channel, variation in canopy height) or indirectly, by applying the Random Forest classification algorithm to the point clouds. Results highlight the potential of 3D LiDAR point clouds to produce indicators with sufficient accuracy (ca. 80% for the number of trunks and 68% for species composition). Our results also reveal advantages of using metrics related to the internal structure of trees, such as penetration indexes. However, intensity metrics calculated using bispectral properties of LiDAR did not improve the quality of classifications. Longitudinal analysis of the indicators revealed a difference in attributes between the reservoir and areas downstream from it.

Combined Lane Mapping Using a Mobile Mapping System

High-definition mapping of 3D lane lines has been widely needed for the highway documentation and intelligent navigation of autonomous systems. A mobile mapping system (MMS) captures both accurate 3D LiDAR point clouds and high-resolution images of lane markings at highway driving speeds, providing an abundant data source for combined lane mapping. This paper aims to map lanes with an MMS. The main contributions of this paper include the following: (1) an intensity correction method was introduced to eliminate the reflectivity inconsistency of road-surface LiDAR points; (2) a self-adaptive thresholding method was developed to extract lane markings from their complicated surroundings; and (3) a LiDAR-guided textural saliency analysis of MMS images was proposed to improve the robustness of lane mapping. The proposed method was tested with a dataset acquired in Wuhan, Hubei, China, which contained straight roads, curved roads, and a roundabout with various pavement markings and a complex roadside environment. The experimental results achieved a recall of 96.4%, a precision of 97.6%, and an F-score of 97.0%, demonstrating that the proposed method has strong mapping ability for various urban roads.

A preprocessing method of 3D point clouds registration in urban environments

Aiming at the problem that 3D LiDAR point cloud has high data density, outlier noise, and scattered distribution in urban environment, which is not conducive to the matching between point clouds in the later stage, a pre-processing method for large-scale LiDAR point cloud frame matching in urban environments is proposed. First, the point cloud data is transformed into a Mean Elevation Map, and the ground point segmentation processing is performed on the point cloud using the height gradient between the grids; then, the DBSCAN clustering algorithm is improved by the three-dimensional voxel grid division method, and the improved VG-DBSCAN is used to cluster point clouds and separate the target point cloud from the outliers after clustering, thereby, which eliminates outlier noises in the point cloud. Finally, the Voxel Grid filter is used to down sample the point cloud. The experimental results show that the proposed method can perform real-time preprocessing on point cloud data, and the average time is 132.1 ms. After pre-processing, the accuracy of point cloud frame matching is increased by 2 times, and the average time consumption is only 1/6 before pre-processing.

DATA MINING FOR CLASSIFICATION OF HIGH VOLUME DENSE LiDAR DATA IN AN URBAN AREA

Abstract. 3D LiDAR point cloud obtained from the laser scanner is too dense and contains millions of points with information. For such huge volume of data to be sorted, identified, validated and be used for prediction, data mining provides immense scope and has been used to achieve the same. Certain unique attributes were selected as an input for creating models through machine learning. Supervised models were thus built for prediction of classes through the available LiDAR data using random forest algorithm. The algorithm was chosen owing to its efficiency and accuracy over other data mining algorithms. The models created using random forest were then tested on an unclassified point cloud data of an urban area. The method shows promising results in terms of classification accuracy as overall accuracy of 91.71 % was achieved for pixel-based classification. The method also displays enhanced efficiency over common classification algorithms as the time taken to make predictions about the data is reduced considerably for a set of dense LiDAR data. This shows positive foresight of making use of data mining and machine learning to handle large volume of LiDAR data and can go a long way in augmenting efficient processing of LiDAR data.

3D Point Cloud Recognition Based on a Multi-View Convolutional Neural Network.

The recognition of three-dimensional (3D) lidar (light detection and ranging) point clouds remains a significant issue in point cloud processing. Traditional point cloud recognition employs the 3D point clouds from the whole object. Nevertheless, the lidar data is a collection of two-and-a-half-dimensional (2.5D) point clouds (each 2.5D point cloud comes from a single view) obtained by scanning the object within a certain field angle by lidar. To deal with this problem, we initially propose a novel representation which expresses 3D point clouds using 2.5D point clouds from multiple views and then we generate multi-view 2.5D point cloud data based on the Point Cloud Library (PCL). Subsequently, we design an effective recognition model based on a multi-view convolutional neural network. The model directly acts on the raw 2.5D point clouds from all views and learns to get a global feature descriptor by fusing the features from all views by the view fusion network. It has been proved that our approach can achieve an excellent recognition performance without any requirement for three-dimensional reconstruction and the preprocessing of point clouds. In conclusion, this paper can effectively solve the recognition problem of lidar point clouds and provide vital practical value.

A Method of Point Cloud Classification Using Multi-scale Dimensionality Features and Transductive Learning

3D point cloud of environments relevant to recognition and monitoring problems in artificial intelligence often require classifying environment data into separated classes. This paper proposes a new classification approach for 3D lidar outdoor environment point clouds. The approach is based on morphological filter, multi-scale dimensionality features and support vector machine model. It can effectively separate the ground, trees and conventional constructions from downsampled point clouds. Compared with single traditional classification algorithm, such as supervoxel clustering algorithm and locally convex connected patches algorithm, the presented approach has good performance in objective and subjective assessments. The effectiveness of the proposed approach is verified based on a public data set collected from the International Society for Photogrammetry and Remote Sensing (ISPRS) and Stanford Modified Object Segmentation Database (MOSD).

Global Registration of 3D LiDAR Point Clouds Based on Scene Features: Application to Structured Environments

Acquiring 3D data with LiDAR systems involves scanning multiple scenes from different points of view. In actual systems, the ICP algorithm (Iterative Closest Point) is commonly used to register the acquired point clouds together to form a unique one. However, this method faces local minima issues and often needs a coarse initial alignment to converge to the optimum. This paper develops a new method for registration adapted to indoor environments and based on structure priors of such scenes. Our method works without odometric data or physical targets. The rotation and translation of the rigid transformation are computed separately, using, respectively, the Gaussian image of the point clouds and a correlation of histograms. To evaluate our algorithm on challenging registration cases, two datasets were acquired and are available for comparison with other methods online. The evaluation of our algorithm on four datasets against six existing methods shows that the proposed method is more robust against sampling and scene complexity. Moreover, the time performances enable a real-time implementation.

THE RESEARCH ON ACCURACY EVALUATION METHOD OF THE TOMOGRAPHIC SAR THREE-DIMENSIONAL RECONSTRUCTION OF URBAN BUILDING BASED ON TERRESTRIAL LIDAR POINT CLOUD

Abstract. Muti-baseLine SAR tomography can be used on 3D reconstruction of urban building based on SAR images acquired. In the near future, it is expected to become an important technical tool for urban multi-dimensional precision monitoring. For the moment，There is no effective method to verify the accuracy of tomographic SAR 3D point cloud of urban buildings. In this paper, a new method based on terrestrial Lidar 3D point cloud data to verify the accuracy of the tomographic SAR 3D point cloud data is proposed, 3D point cloud of two can be segmented into different facadeds. Then facet boundary extraction is carried out one by one, to evaluate the accuracy of tomographic SAR 3D point cloud of urban buildings. The experience select data of Pangu Plaza to analyze and compare, the result of experience show that the proposed method that evaluating the accuracy of tomographic SAR 3D point clou of urban building based on lidar 3D point cloud is validity and applicability

The Potential of Active Contour Models in Extracting Road Edges from Mobile Laser Scanning Data

Active contour models present a robust segmentation approach, which makes efficient use of specific information about objects in the input data rather than processing all of the data. They have been widely-used in many applications, including image segmentation, object boundary localisation, motion tracking, shape modelling, stereo matching and object reconstruction. In this paper, we investigate the potential of active contour models in extracting road edges from Mobile Laser Scanning (MLS) data. The categorisation of active contours based on their mathematical representation and implementation is discussed in detail. We discuss an integrated version in which active contour models are combined to overcome their limitations. We review various active contour-based methodologies, which have been developed to extract road features from LiDAR and digital imaging datasets. We present a case study in which an integrated version of active contour models is applied to extract road edges from MLS dataset. An accurate extraction of left and right edges from the tested road section validates the use of active contour models. The present study provides valuable insight into the potential of active contours for extracting roads from 3D LiDAR point cloud data.

WEAKLY SUPERVISED SEGMENTATION-AIDED CLASSIFICATION OF URBAN SCENES FROM 3D LIDAR POINT CLOUDS

Abstract. We consider the problem of the semantic classification of 3D LiDAR point clouds obtained from urban scenes when the training set is limited. We propose a non-parametric segmentation model for urban scenes composed of anthropic objects of simple shapes, partionning the scene into geometrically-homogeneous segments which size is determined by the local complexity. This segmentation can be integrated into a conditional random field classifier (CRF) in order to capture the high-level structure of the scene. For each cluster, this allows us to aggregate the noisy predictions of a weakly-supervised classifier to produce a higher confidence data term. We demonstrate the improvement provided by our method over two publicly-available large-scale data sets.