Search

Although point features have shown their usefulness in classification with Machine Learning, point cloud visualization enhancement methods focus mainly on lighting. The visualization of point features helps to improve the perception of the 3D environment. This paper proposes Multi Feature-Rich Synthetic Colour (MFRSC) as an alternative non-photorealistic colour approach of natural-coloured point clouds. The method is based on the selection of nine features (reflectance, return number, inclination, depth, height, point density, linearity, planarity, and scattering) associated with five human perception descriptors (edges, texture, shape, size, depth, orientation). The features are reduced to fit the RGB display channels. All feature permutations are analysed according to colour distance with the natural-coloured point cloud and Image Quality Assessment. As a result, the selected feature permutations allow a clear visualization of the scene's rendering objects, highlighting edges, planes, and volumetric objects. MFRSC effectively replaces natural colour, even with less distorted visualization according to BRISQUE, NIQUE and PIQE. In addition, the assignment of features in RGB channels enables the use of MFRSC in software that does not support colorization based on point attributes (most commercially available software). MFRSC can be combined with other non-photorealistic techniques such as Eye-Dome Lighting or Ambient Occlusion. [ABSTRACT FROM AUTHOR]

To reduce the cost of manually annotating training data for supervised classifiers, we propose an automated approach to extract training data of urban objects in six classes: buildings, fences, man-made poles, vegetation, vehicles, and low objects. In this study, two segmentation algorithms are firstly implemented to generate meaningful objects from the non-ground point cloud. Then, we generated valid strict rules to label partial RANSAC (Random Sample Consensus) planes and meaningful objects as training data. The strict rules are built upon the semantic knowledge formed by the features of geometric, eigenvalue, RANSAC plane, multidimensional slice, and relative location. The accuracy of strict rule-based (SRB) training data is higher than 98.5 % for buildings, man-made poles, vegetation, and vehicles. The accuracy of low objects and fences reaches 97.10 % and 94.99 %, respectively. Finally, we compared the performance of the KPConv and PointNET++ networks trained by SRB and manually labeled training data to evaluate the effectiveness of our training data. The KPConv overall accuracy using manually labeled and (SRB) training data are 91.5 % and 86.8 % in the Paris dataset, 95.6 % and 92.0 % in the Freiburg dataset, respectively. The experiments demonstrate that automatically labeled training data can achieve similar accuracy compared to manual labels when coupled with two deep learning networks. Therefore, SRB training data extraction can effectively deal with the problem of training data scarcity and provide significant advancements in urban point cloud classification, where manual labeling of training data remains a crucial challenge. [ABSTRACT FROM AUTHOR]

[Display omitted] Deep learning (DL) semantic segmentation of tunnel point cloud shows an efficient path for applications related to subway tunnel scenes, such as health inspection and building information modelling (BIM). Current methods for tunnel point cloud segmentation often suffer from a shortage of benchmarks. This paper proposed a large-scale, multi-modal dataset for semantic segmentation of subway tunnel point cloud called subway tunnel segmentation dataset (STSD). The STSD comprises point clouds and projected images annotated into 12 categories, encompassing three types of subway tunnels with a combined length exceeding 2700 m, totaling over 2.26 billion points. A novel approach for DL semantic segmentation of subway tunnel point clouds is proposed herein. This approach enables the direct utilization of image-based DL segmentation networks on subway tunnel point clouds. Furthermore, it incorporates a lossless coordinate transformation method capable of converting tunnel point clouds of any cross-section shape into images with minimal information loss. Further evaluation of several classic or state-of-the-art 2D and 3D DL semantic segmentation models shows the feasibility of the approach and dataset. The best 2D model achieves a mIoU of 86.26% and outperforms the best 3D model by almost 10%. This research provides a novel approach for DL semantic segmentation in subway tunnel point clouds, contributes a large-scale, multi-modal dataset for the tunnel semantic segmentation, and creates a benchmark for further evaluation of the corresponding algorithms. [ABSTRACT FROM AUTHOR]

Outdated or sketchy inventory of street furniture may misguide the planners on the renovation and upgrade of transportation infrastructures, thus posing potential threats to traffic safety. Previous studies have taken their steps using point clouds or street-view imagery (SVI) for street furniture inventory, but there remains a gap to balance semantic richness, localization accuracy and working efficiency. Therefore, this paper proposes an effective pipeline that combines SVI and point clouds for the inventory of street furniture. The proposed pipeline encompasses three steps: (1) Off-the-shelf street furniture detection models are applied on SVI for generating two-dimensional (2D) proposals and then three-dimensional (3D) point cloud frustums are accordingly cropped; (2) The instance mask and the instance 3D bounding box are predicted for each frustum using a multi-task neural network; (3) Frustums from adjacent perspectives are associated and fused via multi-object tracking, after which the object-centric instance segmentation outputs the final street furniture with 3D locations and semantic labels. This pipeline was validated on datasets collected in Shanghai and Wuhan, producing component-level street furniture inventory of nine classes. The instance-level mean recall and precision reach 86.4%, 80.9% and 83.2%, 87.8% respectively in Shanghai and Wuhan, and the point-level mean recall, precision, weighted coverage all exceed 73.7%. [ABSTRACT FROM AUTHOR]

• Proposed a new method for extracting and segmenting street tree point clouds. • Accurately identifying tree trunks based on PCA and layered circle fitting. • The edge point cloud segmentation accuracy of overlapping trees is more robust. • Developed a stable technical solution from preprocessing to parameter acquisition. Trees are a crucial component of urban environments. However, accurately identifying and segmenting individual street trees from Mobile Laser Scanning (MLS) point clouds in complex urban scenes remains a significant challenge due to low accuracy and poor robustness. Additionally, traditional manual extraction methods are labor-intensive and time-consuming. To address these issues, this paper proposes a method for the automatic extraction and segmentation of urban street trees using MLS point cloud data. The preprocessing phase initially filters out ground and building points to reduce computational complexity. Subsequently, coarse extraction is performed using projection density, and a method based on spatial morphology and elevation slice frequency statistics is employed to differentiate other objects and cluster street trees. Next, the Principal Component Analysis (PCA) algorithm is employed to initialize the potential tree trunk point cloud, and precise identification of tree trunks is achieved through the application of circular arc curvature and a layered circular fitting model. Finally, a combination of neighborhood-based connectivity coefficients and a shortest path constraint method is utilized to accurately segment multiple overlapping street trees. Evaluation results from multiple sets of experimental datasets indicate that the proposed algorithm achieves a high level of accuracy in trunk extraction, resulting in clearer outlines of the segmented tree crowns. The detection accuracy and precision of individual tree clusters are approximately 95.3% and 96.7%, respectively, presenting more robust segmentation results compared to mainstream methods. This contributes significantly to providing crucial technical references for urban tree census and statistics. [ABSTRACT FROM AUTHOR]

Accidents and fatalities during scaffolding work are significant concerns in the construction industry. Scaffold monitoring technology based on 3D reconstruction can offer a potential solution to prevent such fatalities. However, the practical application of this technology at actual construction sites faces a notable challenge due to the presence of safety nets. Safety nets hinder the acquisition of complete scaffold point clouds necessary for reconstruction. To address this issue, we propose a novel registration method for scaffold point clouds that considers safety nets, employing sensor fusion and deep learning techniques. The method consists of three steps: 1) LiDAR-camera calibration and SLAM-based point cloud data acquisition, 2) object detection and coordinate transformation, and 3) identification of scaffold installation stages and point cloud registration. The proposed method was verified to be effective in registering scaffold point clouds without safety nets. • The proposed method registers scaffold point clouds considering the installation of safety nets for 3D reconstruction. • Sensor fusion and deep learning are used to estimate the global coordinates of objects during scaffolding work. • By identifying the installation stages of scaffolds, a point cloud map without safety nets is obtained. • The effectiveness of the method is validated at the real scaffold installation site. [ABSTRACT FROM AUTHOR]

Multipath effects and signal obstruction by buildings in urban canyons can lead to inaccurate GNSS measurements and therefore errors in the estimated trajectory of Mobile Laser Scanning (MLS) systems; consequently, derived point clouds are distorted and lose spatial consistency. We obtain decimetre-level trajectory accuracy making use of corresponding points between the MLS data and aerial images with accurate exterior orientations instead of using ground control points. The MLS trajectory is estimated based on observation equations resulting from these corresponding points, the original IMU observations, and soft constraints on the pitch and yaw rotations of the vehicle. We analyse the quality of the trajectory enhancement under several conditions where the experiments were designed to test the influence of the number and quality of corresponding points and to test different settings for a B-spline representation of the vehicle trajectory. The method was tested on two independently acquired MLS datasets in Rotterdam by enhancing the trajectories and evaluating them using checkpoints. The RMSE values of the original GNSS/IMU based Kalman filter results at the checkpoints were 0.26 m, 0.30 m, and 0.47 m for the X-, Y- and Z-coordinates in the first dataset and 1.10 m, 1.51 m, and 1.81 m in the second dataset. The latter dataset was recorded with a lower quality IMU in an area with taller buildings. After trajectory adjustment these RMSE values were reduced to 0.09 m, 0.11 m, and 0.16 m for the first dataset and 0.12 m, 0.14 m, and 0.18 m for the second dataset. The results confirmed that, if sufficient tie points between the point cloud and aerial imagery are available, the method supports geo-referencing of MLS point clouds in urban canyons with a near-decimetre accuracy. [ABSTRACT FROM AUTHOR]

Forest inventories rely on field plots, the measurement of which is costly and time consuming by manual means. Thus, there is a need to automate plot-level field data collection. Mobile laser scanning has yet to be demonstrated for deriving stem curve and volume from standing trees with sufficient accuracy for supporting forest inventory needs. We tested a new approach based on pulse-based backpack mobile laser scanner (Riegl VUX-1HA) combined with in-house developed SLAM (Simultaneous Localization and Mapping), and a novel post-processing algorithm chain that allows one to extract stem curves from scan-line arcs corresponding to individual standing trees. The post-processing step included, among others, an algorithm for scan-line arc extraction, a stem inclination angle correction and an arc matching algorithm correcting for the drifts that are still present in the stem points after applying the SLAM algorithm. By using the stem curves defined by the detected arcs and tree heights provided by the pulse-based scanner, stem volume estimates for standing trees in easy (n = 40) and medium (n = 37) difficult boreal forest were calculated. In the easy and medium plots, 100% of pine and birch stems were correctly detected. The total RMSE of the extracted stem curves was 1.2 cm (5.1%) and 1.7 cm (6.7%) for the easy and medium plots, respectively. The RMSE were 1.8 m (8.7%) and 1.1 m (4.9%) for the estimated tree heights, and 9.7% and 10.9% for the stem volumes for the easy and medium plots, correspondingly. Thus, our processing chain provided stem volume estimates with a better accuracy than previous methods based on mobile laser scanning data. Importantly, the accuracy of stem volume estimation was comparable to that provided by terrestrial laser scanning approaches in similar forest conditions. To further demonstrate the performance of the proposed method, we compared our results against stem volumes calculated using the standard Finnish allometric volume model, and found that our method provided more accurate volume estimates for the two test sites. The findings are important steps towards future individual-tree-based airborne laser scanning inventories which currently lack cost-efficient and accurate field reference data collection techniques. The tree geometry defined by the stem curve is also an important input parameter for deriving quality-related information from trees. Forest management decision making will benefit from improvements to the efficiency and quality of individual tree reference information. [ABSTRACT FROM AUTHOR]

Falls from scaffolds cause the majority of accidents and fatalities at construction sites. A deep learning-based 3D reconstruction technology could provide a solution to prevent such fatalities through automated scaffold monitoring. However, when the technology was used at a large-scale construction site, there were limitations, such as the scarcity of point cloud data and the non-uniformity of points. To address this issue, this paper presents a large-scale scaffold reconstruction method using synthetic scaffold datasets and an upsampling adversarial network. The method consists of four steps: 1) data acquisition of scaffold point cloud through a mobile laser scanning (MLS) system, 2) 3D semantic segmentation using synthetic datasets, 3) upsampling of the segmented scaffold points, and 4) automatic generation of a 3D CAD model. The performance of the segmentation model trained with synthetic datasets achieved an 80.83% F1 score, which improved to 94.93% after upsampling. • The proposed method reconstructs a large-scale scaffold using a MLS system. • A synthetic dataset generation algorithm was developed to overcome data scarcity. • Scaffold points became dense after using an upsampling adversarial network. • An F1 score for scaffold point clouds was 94.93%. [ABSTRACT FROM AUTHOR]

The mapping of road boundaries provides critical information about roads for urban road traffic safety. This paper presents a deep learning-based framework for recovering 3D road boundary using multi-source data, which include mobile laser scanning (MLS) point clouds, spatial trajectory data, and remote sensing images. The proposed road recovery method uses extracted 3D road boundaries from MLS point clouds as inputs. First, after automatic erroneous boundary removal, a CNN-based boundary completion model completes road boundaries. Then, to refine the imperfect road boundaries, road centerlines generated from dynamic taxi GPS trajectory data and remote sensing images are used as completion guidance for a generative adversarial nets model to obtain more accurate and complete road boundaries. Finally, after associating a sequence of taxi GPS recorded trajectory points with the correct 3D road boundaries, inherent geometric road characteristics and road dynamic information are extracted from the complete boundaries and taxi GPS trajectory data, respectively. The testing dataset contains two urban road MLS datasets, and the KITTI dataset. The experimental results on point clouds from different sensors demonstrate our proposed method is effective and promising for recovering 3D road boundary and extracting road characteristics. [ABSTRACT FROM AUTHOR]

Road furniture recognition has become a prevalent issue in the past few years because of its great importance in smart cities and autonomous driving. Previous research has especially focussed on pole-like road furniture, such as traffic signs and lamp posts. Published methods have mainly classified road furniture as individual objects. However, most road furniture consists of a combination of classes, such as a traffic sign mounted on a street light pole. To tackle this problem, we propose a framework to interpret road furniture at a more detailed level. Instead of being interpreted as single objects, mobile laser scanning data of road furniture is decomposed in elements individually labelled as poles, and objects attached to them, such as, street lights, traffic signs and traffic lights. In our framework, we first detect road furniture from unorganised mobile laser scanning point clouds. Then detected road furniture is decomposed into poles and attachments (e.g. traffic signs). In the interpretation stage, we extract a set of features to classify the attachments by utilising a knowledge-driven method and four representative types of machine learning classifiers, which are random forest, support vector machine, Gaussian mixture model and naïve Bayes, to explore the optimal method. The designed features are the unary features of attachments and the spatial relations between poles and their attachments. Two experimental test sites in Enschede dataset and Saunalahti dataset were applied, and Saunalahti dataset was collected in two different epochs. In the experimental results, the random forest classifier outperforms the other methods, and the overall accuracy acquired is higher than 80% in Enschede test site and higher than 90% in both Saunalahti epochs. The designed features play an important role in the interpretation of road furniture. The results of two epochs in the same area prove the high reliability of our framework and demonstrate that our method achieves good transferability with an accuracy over 90% through employing the training data of one epoch to test the data in another epoch. [ABSTRACT FROM AUTHOR]

Poor GNSS measurements in urban areas caused by blocked GNSS signals and multi-path is a well-known problem, which leads to an inaccurate trajectory estimation of Mobile Laser Scanning (MLS) platforms. Consequently, the MLS point cloud contains positioning errors. This paper presents a new method for the automatic extraction of accurate 3D tie points for the trajectory adjustment of MLS platforms in GNSS denied or troubled areas. The new method relies on aerial imagery as a reliable external source of reference provided that accurate exterior orientation parameters are available. Accordingly, one of the main objectives is to register the mobile laser scanning point cloud with corresponding aerial images. The matches between aerial images are used to obtain 3D tie points by forward intersection. By also determining the corresponding locations in the point cloud we obtain a 3D-3D correspondence between the MLS point cloud and the aerial images. In the future, the obtained 3D-3D correspondences will be used for trajectory adjustment. Our automatic tie point extraction procedure is tested on two independent MLS point clouds. The point clouds were acquired by two different platforms in Rotterdam. The aerial imagery of the same area was acquired at a different time. We evaluated the matching results for both datasets and concluded that the new procedure reliably extracted the 3D tie points for 55% of the tiles of the size of 90 m from the first MLS dataset. In the second dataset, 60% of the tiles of size 74 m yielded reliable 3D tie points. It is not necessary to successfully register all tiles because the results of this work will be used for the trajectory adjustment and the IMU can reliably support the positioning for small intervals. [ABSTRACT FROM AUTHOR]

Objects are formed by various structures and such structural information is essential for the identification of objects, especially for street facilities presented by mobile laser scanning (MLS) data with abundant details. However, due to the large volume of data, large variations in point density, noise and complexity of scanned scenes, the achievement of effective decomposition of objects into physical meaningful structures remains a challenge issue. And structural information has been rarely considered to improve the accuracy of distinguishing between objects with global or local similarity, such as traffic signs and traffic lights. Therefore, we propose a structural segmentation and classification method for MLS point clouds that is efficient and robust to variations in point density and complex urban scenes. During the segmentation stage, a novel region growing approach and a multi-size supervoxel segmentation algorithm robust to noise and varying density are combined to extract effective local shape descriptors. Structural components with physically meaningful labels are generated via structural labelling and clustering. During the classification stage, we consider the structural information at various scales and locations and encode it into a conditional random-field model for unary and pairwise inferences. High-order potentials are also introduced into the conditional random field to eliminate regional label noise. These high-order potentials are defined upon regions independent of connection relationships and can therefore take effect on isolated nodes. Experiments with two MLS datasets of typical urban scenes in Paris and Hong Kong were used to evaluate the performance of the proposed method. Nine and eleven different object classes were recognized from these two datasets with overall accuracies of 97.13% and 95.79%, respectively, indicating the effectiveness of the proposed method of interpreting complex urban scenes from point clouds with large variations in point density. Compared with previous studies on the Paris dataset, our method was able to recognize more classes and obtained a mean F 1 -score of 72.70% of seven common classes, being higher than the best of previous results. [ABSTRACT FROM AUTHOR]

Plane detection and segmentation is one of the most crucial tasks in point cloud processing. The output from this process can be used as input for further processing steps, such as modelling, registration and calibration. However, the sparseness and heterogeneity of Mobile Laser Scanning (MLS) point clouds may lead to problems for existing planar surfaces detection and segmentation methods. This paper proposes a new method that can be applicable to detect and segment planar features in sparse and heterogeneous MLS point clouds. This method utilises the scan profile patterns and the planarity values between different neighbouring scan profiles to detect and segment planar surfaces from MLS point clouds. The proposed method is compared to the three most state-of-the-art segmentation methods (e.g. RANSAC, a robust segmentation method based on robust statistics and diagnostic principal component analysis – RDCPA as well as the plane detection method based on line arrangement). Three datasets are used for the validation of the results. The results show that our proposed method outperforms the existing methods in detecting and segmenting planar surfaces in sparse and heterogeneous MLS point clouds. In some instances, the state-of-the-art methods produce incorrect segmentation results for façade details which have a similar orientation, such as for windows and doors within a façade. While RDCPA produces up to 50% of outliers depending on the neighbourhood threshold, another method could not detect such features at all. When dealing with small features such as a target, some algorithms (including RANSAC) were unable to perform segmentation. However, the propose algorithm was demonstrated to detect all planes in the test data sets correctly. The paper shows that these mis-segmentations in other algorithms may lead to significant errors in the registration process of between 1.047 and 1.614 degrees in the angular parameters, whereas the propose method had only resulted in 0.462 degree angular bias. Furthermore, it is not sensitive to the required method parameters as well as the point density of the point clouds. [ABSTRACT FROM AUTHOR]

Highlights • Direct use of point clouds to generate navigable graphs for pedestrians in urban environments. • Real routes generated according to two motor skills: pedestrians without reduced mobility and wheelchairs. • Navigable space based on ground elements and static objects. • Occlusion correction of sidewalks by applying morphological operations. Abstract Pathfinding applications for the citizen in urban environments are usually designed from the perspective of a driver, not being effective for pedestrians. In addition, urban scenes have multiple elements that interfere with pedestrian routes and navigable space. In this paper, a methodology for the direct use of point clouds for pathfinding in urban environments is presented, solving the main limitations for this purpose: (a) the excessive number of points is reduced for transformation into nodes on the final graph, (b) urban static elements acting as permanent obstacles, such as furniture and trees, are delimited and differentiated from dynamic elements such as pedestrians, (c) occlusions on ground elements are corrected to enable a complete graph modelling, and (d) navigable space is delimited from free unobstructed space according to two motor skills (pedestrians without reduced mobility and wheelchairs). The methodology is tested into three different streets sampled as point clouds by mobile laser scanning (MLS) systems: an intersection of several streets with ground composed of sidewalks at different heights; an avenue with wide sidewalks, trees and cars parked on one side; and a street with a single-lane road and narrow sidewalks. By applying Dijkstra pathfinding algorithm to the resulting graphs, the correct viability of the generated routes has been verified based on a visual analysis of the generated routes on the point cloud and on the knowledge of the urban study area. The methodology enables the automatic generation of graphs representing the navigable urban space, on which safe and real routes for different motor skills can be calculated. [ABSTRACT FROM AUTHOR]

Abstract Surveys of roadways with Mobile Laser Scanning (MLS) are now being conducted on a regular basis by many transportation agencies to provide detailed geometric information to support a wide range of applications, including asset management. Most MLS systems provide intensity (return signal strength) data as a point attribute in georeferenced point clouds, which may be used to estimate retro -reflectivity of pavement markings for effective maintenance. Nevertheless, the extraction of pavement markings from mobile lidar data remains an open challenge, due to variable noise, degree of wear on the markings, and road conditions. This paper addresses these challenges, presenting a novel approach for efficient, reliable extraction of lane markings, including those that have been significantly worn. First, using the MLS trajectory information, the lidar data is discretized into smaller sections, and then transformed to the local coordinate system, such that the road surface is near-horizontal for reliable extraction on roads with significant grade. Subsequently, the road surface is extracted using the constrained Random Sampling and Consensus (RANSAC) algorithm and then rasterized into a 2D intensity image to apply image processing techniques, namely: image segmentation to separate the lane markings from the road pavement, and a morphological opening operation to remove small objects. However, the extracted lane markings are prone to over-segmentation, due to occlusions or worn portions caused by moving vehicles. To rectify this, topologically-similar lane markings are associated with each other by computing line parameters (i.e., orientation and distance from the origin), which enables the gaps to be filled among the associated lanes. Finally, the remaining incorrect lane markings are detected and removed through a noise filtering phase using Dip test statistics. Examples of the effectiveness and application of the methodology are shown for a variety of sites with stripes of variable condition to highlight the robustness of the approach. Using optimized parameter values, the algorithm achieved F1 scores of 89–97% when tested on a variety of datasets encompassing a wide range of road scene types. [ABSTRACT FROM AUTHOR]

Accurate assessments of forest resources rely on ground truth data that are collected via in-situ measurements, which are fundamental for all other statistical- and/or remote-sensing-based deductions on quantified forest attributes. The major bottleneck of the current in-situ observation system is that the data collection is time consuming, and, thus, limited in extent, which potentially biases any further inferences made. Consequently, conventional field-data-collection approaches can hardly keep pace with the coverage, scale and frequency required for contemporary and future forest inventories. In-situ measurements from mobile platforms seem to be a promising technique to solve this problem and are estimated at least 10 times faster than static techniques (e.g., terrestrial laser scanning, TLS) at the plot level. However, the mobile platforms are still at the very early stages of development, and it is unclear which three-dimensional (3D) forest measurements the mobile systems can provide and at what accuracy. This study presents a quantitative evaluation of the performance of mobile platforms in a variety of forest conditions and through a comparison with state-of-the-art static in-situ observations. Two mobile platforms were used to collect field data, where the same laser-scanning system was both mounted on top of a vehicle and wore by an operator. The static in-situ observation from TLS is used as a baseline for the evaluation. All point clouds involved were processed through the same processing chain and compared to conventional manual measurement. The evaluation results indicate that the mobile platforms can assess homogeneous forests as well as static observations, but they cannot yet assess heterogeneous forest as required by practical applications. The major challenge is twofold: mobile-data coverage and accuracy. Future research should focus on the robust registration techniques between strips, especially in complex forest conditions, since errors of data registration results in significant impacts on tree attributes estimation accuracy. In cases that the spatial inconstancy cannot be eliminated, attributes estimation in single strips, i.e., the multi-single-scan approach, is an alternative. Meanwhile, operator training deserves attention since the data quality from mobile platforms is partly determined by the operators’ selection of trajectory in the field. [ABSTRACT FROM AUTHOR]

Indoor building models are essential in many indoor applications. These models are composed of the primitives of the buildings, such as the ceilings, floors, walls, windows, and doors, but not the movable objects in the indoor spaces, such as furniture. This paper presents, for indoor environments, a novel semantic line framework-based modeling building method using backpacked laser scanning point cloud data. The proposed method first semantically labels the raw point clouds into the walls, ceiling, floor, and other objects. Then line structures are extracted from the labeled points to achieve an initial description of the building line framework. To optimize the detected line structures caused by furniture occlusion, a conditional Generative Adversarial Nets (cGAN) deep learning model is constructed. The line framework optimization model includes structure completion, extrusion removal, and regularization. The result of optimization is also derived from a quality evaluation of the point cloud. Thus, the data collection and building model representation become a united task-driven loop. The proposed method eventually outputs a semantic line framework model and provides a layout for the interior of the building. Experiments show that the proposed method effectively extracts the line framework from different indoor scenes. [ABSTRACT FROM AUTHOR]

In the framework of infrastructure analysis and maintenance in an urban environment, it is important to address the safety of every road user. This paper presents a methodology for the evaluation of several safety indicators on pedestrian crossing environments using geometric and radiometric information extracted from 3D point clouds collected by a Mobile Mapping System (MMS). The methodology is divided in four main modules which analyze the accessibility of the crossing area, the presence of traffic lights and traffic signs, and the visibility between a driver and a pedestrian on the proximities of a pedestrian crossing. The outputs of the analysis are exported to a Geographic Information System (GIS) where they are visualized and can be further processed in the context of city management. The methodology has been tested on approximately 30 pedestrian crossings in cluttered urban environments of two different cities. Results show that MMS are a valid mean to assess the safety of a specific urban environment, regarding its geometric conditions. Remarkable results are presented on traffic light classification, with a global F-score close to 95%. [ABSTRACT FROM AUTHOR]

In this study we have investigated why diameters of tree stems, which are approximately cylindrical, are often overestimated by mobile laser scanning. This paper analyzes the physical processes when using ground-based laser scanning that may contribute to a bias when estimating cylinder diameters using circle-fit methods. A laser scanner simulator was implemented and used to evaluate various properties, such as distance, cylinder diameter, and beam width of a laser scanner-cylinder system to find critical conditions. The simulation results suggest that a positive bias of the diameter estimation is expected. Furthermore, the bias follows a quadratic function of one parameter – the relative footprint, i.e., the fraction of the cylinder width illuminated by the laser beam. The quadratic signature opens up a possibility to construct a compensation model for the bias. [ABSTRACT FROM AUTHOR]

• Novel multi-view based deep learning approach for colored point cloud completion. • Generation of dense point clouds with a variable number of points. • Preserving the genuine high-level detailedness of laser scanning measurements. • A novel late fusion-based technique for integrating the view-level features. • New public Benchmark containing both synthetic point clouds and real MLS data. In this paper, we introduce a novel multi view-based method for completing high-resolution 3D point clouds of partial object shapes obtained by mobile laser scanning (MLS) platforms. Our approach estimates both the geometry and color cues of the missing or incomplete object segments, by projecting the 3D input point cloud by multiple virtual cameras, and performing 2D inpainting in the image domains of the different views. In contrast to existing state-of-the-art methods, our method can generate point clouds consisting of a variable number of points, depending on the detailedness of the input measurement, which property highly facilitates the efficient processing of MLS data with inhomogeneous point density. For training and quantitative evaluation of the proposed method, we provide a new point cloud dataset that consists of both synthetic point clouds of four different street objects with accurate ground truth, and real MLS measurements of partially or fully scanned vehicles. The quantitative and qualitative experiments on the provided dataset demonstrate that our method surpasses state-of-the-art approaches in reconstructing the local fine geometric structures as well as in estimating the overall shape and color pattern of the objects. [ABSTRACT FROM AUTHOR]

Building accessibility diagnosis is of high interest especially in case of people with reduced mobility. This paper proposes a methodology for automated detection of inaccessible steps in building façade entrances from MLS (mobile laser scanner) data. Our approach uses the MLS trajectory to automatically subdivide urban point clouds into regular stretches. From each stretch, the lower zone of façade is isolated and selected as region of interest. Points belonging to vertical elements are projected onto a 2D image and steps are detected and classified as inaccessible areas according to the comparison of geometrical features such as height jump, proximity to ground and width, with regulation. The methodology has been tested in four real datasets, which constitute > 400 m of different urban scenarios. Results exhibit a robust performance under urban scenes with a high variability of façade geometry due to the presence of different entrance types to shops and dwellings. Results have been quantitatively evaluated and they show global F1 value around 93%. Moreover, the methodology is very fast since 100 m are processed in < 2 min. [ABSTRACT FROM AUTHOR]

Mobile laser scanning (MLS) provides kinematic means to collect three dimensional data from surroundings for various mapping and environmental analysis purposes. Vehicle based MLS has been used for road and urban asset surveys for about a decade. The equipment to derive the trajectory information for the point cloud generation from the laser data is almost without exception based on GNSS-IMU (Global Navigation Satellite System – Inertial Measurement Unit) technique. That is because of the GNSS ability to maintain global accuracy, and IMU to produce the attitude information needed to orientate the laser scanning and imaging sensor data. However, there are known challenges in maintaining accurate positioning when GNSS signal is weak or even absent over long periods of time. The duration of the signal loss affects the severity of degradation of the positioning solution depending on the quality/performance level of the IMU in use. The situation could be improved to a certain extent with higher performance IMUs, but increasing system expenses make such approach unsustainable in general. Another way to tackle the problem is to attach additional sensors to the system to overcome the degrading position accuracy: such that observe features from the environment to solve for short term system movements accurately enough to prevent the IMU solution to drift. This results in more complex system integration with need for more calibration and synchronization of multiple sensors into an operational approach. In this paper we study operation of an ATV (All -terrain vehicle) mounted, GNSS-IMU based single scanner MLS system in boreal forest conditions. The data generated by RoamerR2 system is targeted for generating 3D terrain and tree maps for optimizing harvester operations and forest inventory purposes at individual tree level. We investigate a process-flow and propose a graph optimization based method which uses data from a single scanner MLS for correcting the post-processed GNSS-IMU trajectory for positional drift under mature boreal forest canopy conditions. The result shows that we can improve the internal conformity of the data significantly from 0.7 m to 1 cm based on tree stem feature location data. When the optimization result is compared to reference at plot level we reach down to 6 cm mean error in absolute tree stem locations. The approach can be generalized to any MLS point cloud data, and provides as such a remarkable contribution to harness MLS for practical forestry and high precision terrain and structural modeling in GNSS obstructed environments. [ABSTRACT FROM AUTHOR]

Urban road environments contain a variety of objects including different types of lamp poles and traffic signs. Its monitoring is traditionally conducted by visual inspection, which is time consuming and expensive. Mobile laser scanning (MLS) systems sample the road environment efficiently by acquiring large and accurate point clouds. This work proposes a methodology for urban road object recognition from MLS point clouds. The proposed method uses, for the first time, shape descriptors of complete objects to match repetitive objects in large point clouds. To do so, a novel 3D multi-scale shape descriptor is introduced, that is embedded in a workflow that efficiently and automatically identifies different types of lamp poles and traffic signs. The workflow starts by tiling the raw point clouds along the scanning trajectory and by identifying non-ground points. After voxelization of the non-ground points, connected voxels are clustered to form candidate objects. For automatic recognition of lamp poles and street signs, a 3D significant eigenvector based shape descriptor using voxels (SigVox) is introduced. The 3D SigVox descriptor is constructed by first subdividing the points with an octree into several levels. Next, significant eigenvectors of the points in each voxel are determined by principal component analysis (PCA) and mapped onto the appropriate triangle of a sphere approximating icosahedron. This step is repeated for different scales. By determining the similarity of 3D SigVox descriptors between candidate point clusters and training objects, street furniture is automatically identified. The feasibility and quality of the proposed method is verified on two point clouds obtained in opposite direction of a stretch of road of 4 km. 6 types of lamp pole and 4 types of road sign were selected as objects of interest. Ground truth validation showed that the overall accuracy of the ∼170 automatically recognized objects is approximately 95%. The results demonstrate that the proposed method is able to recognize street furniture in a practical scenario. Remaining difficult cases are touching objects, like a lamp pole close to a tree. [ABSTRACT FROM AUTHOR]

Traffic signs are one of the most important safety elements in a road network. Particularly, road markings provide information about the limits and direction of each road lane, or warn the drivers about potential danger. The optimal condition of road markings contributes to a better road safety. Mobile Laser Scanning technology can be used for infrastructure inspection and specifically for traffic sign detection and inventory. This paper presents a methodology for the detection and semantic characterization of the most common road markings, namely pedestrian crossings and arrows. The 3D point cloud data acquired by a LYNX Mobile Mapper system is filtered in order to isolate reflective points in the road, and each single element is hierarchically classified using Neural Networks. State of the art results are obtained for the extraction and classification of the markings, with F-scores of 94% and 96% respectively. Finally, data from classified markings are exported to a GIS layer and maintenance criteria based on the aforementioned data are proposed. [ABSTRACT FROM AUTHOR]

• A piecewise distance effect function and the Blinn-Phong model-based angle effect correction is proposed to correct the raw intensity for road marking detection. • The background data-driven angle effect function (BDAF) is beneficial for enhancing marking and road classification. • The incidence angle span (IAS) has a major influence on intensity correction of road marking, the road marking separability was improved by 35%@535 pt/m2 when IAS was increased from 3.17° to 13.5°. The intensity data recorded by light detection and ranging (LiDAR) contains the spectral characteristics of a scanned target, making it an essential source of information in mobile laser scanning (MLS) related applications. However, the raw intensity can be distorted by distance and incidence angle, obstructing its use in autonomous driving. To correct the intensity distortion, an automatic demarcation selection, based on distance effect function, and a Blinn-Phong model-based angle effect correction, have been proposed to correct the raw intensity, establishing the relationship between intensity correction performance and the influencing factors, including modeling data selection, incidence angle span (IAS) and point cloud density. Separability (SEP) and balanced F score (BFS) were used to evaluate the separability, accuracy and totality of road marking detection. Experimental results showed that the background data-driven angle effect function (BDAF) used was beneficial for enhancing marking and road classification. IAS has a major influence on correcting intensity performance. SEP was improved by 35% when IAS was increased from 3.17° to 13.5°, with point density of 535 pt/m2. The IAS threshold could be adjusted according to the instrument performance used. It offers a meaningful investigation of detection and classification of planar target using corrected MLS intensity. [ABSTRACT FROM AUTHOR]

• The total leaf area (TLA) of the canopy was evaluated for variable-rate spraying. • The total grid area (TGA) of the canopy was measured by the first and second echoes of a 2D LiDAR. • The TGA-based TLA evaluation method can adapt to varying point density. In recent years, variable-rate spraying technology has been proposed to solve the pesticide waste and environmental pollution problems caused by traditional spraying methods. Variable-rate spraying methods use sensors to measure the leaf amounts in the target canopies, which are used to adjust the amount of liquid and ensure even spraying for canopies of different sizes and densities. One of these methods uses mobile laser scanning (MLS) technologies with a two-dimensional (2D) light detection and ranging (LiDAR) sensor to capture the point cloud of the canopy, and then the total point number (TPN) is used to evaluate the total leaf area (TLA) of the canopy as a basis for variable-rate spraying. However, the moving speed and ranging distance of LiDAR often vary in the process of spraying, resulting in an inconsistent TPN and a large influence on the accuracy of the TLA measurement. In this paper, a new method for estimating the TLA of an individual tree canopy is proposed based on the total grid area (TGA) measured from the first and second echoes of a LiDAR sensor. The grid area at a single point is calculated by multiplying the LiDAR moving resolution and scanning resolution. To evaluate the proposed method, we use an artificial tree, the branches of which are pruned to generate nine canopies with different densities. Both sides of the canopies are scanned at three distances, namely, 1.0 m, 1.5 m, and 2.0 m, and at a constant moving speed of 0.104 m/s. Then, the point cloud of each canopy is sampled multiple times to simulate a variety of moving speeds ranging from 1.040 m/s to 2.080 m/s. Finally, a total of 594 test canopies are obtained. Univariate polynomial regression is adopted to evaluate the TLA by using the TPN and TGA. When the polynomial order is set to 2, the coefficient of determination (R-squared) of the TPN evaluation is 0.5759 with a 4662.4542 cm2 root mean squared error (RMSE), and the R-squared of the TGA evaluation is 0.9617 with a 1401.1652 cm2 RMSE. The experimental results show that the proposed method can effectively eliminate the effects of different LiDAR moving speeds and ranging distances during TLA measurements. [ABSTRACT FROM AUTHOR]

Decision making on forest resources relies on the precise information that is collected using inventory. There are many different kinds of forest inventory techniques that can be applied depending on the goal, scale, resources and the required accuracy. Most of the forest inventories are based on field sample. Therefore, the accuracy of the forest inventories depends on the quality and quantity of the field sample. Conventionally, field sample has been measured using simple tools. When map is required, remote sensing materials are needed. Terrestrial laser scanning (TLS) provides a measurement technique that can acquire millimeter-level of detail from the surrounding area, which allows rapid, automatic and periodical estimates of many important forest inventory attributes. It is expected that TLS will be operationally used in forest inventories as soon as the appropriate software becomes available, best practices become known and general knowledge of these findings becomes more wide spread. Meanwhile, mobile laser scanning, personal laser scanning, and image-based point clouds became capable of capturing similar terrestrial point cloud data as TLS. This paper reviews the advances of applying TLS in forest inventories, discusses its properties with reference to other related techniques and discusses the future prospects of this technique. [ABSTRACT FROM AUTHOR]

Vehicle-related ground occlusion is a common problem in MLS data. This study aims to design a detection and reconstruction method of static vehicle-related ground occlusion for MLS data. Ground extraction and vehicle segmentation are performed on the input point cloud data in advance. Then an α-shape boundary based on the prior vehicle geometry is designed to split non-ground empty area and ground occlusions. The occlusion is detected and matched with its corresponding vehicle using the relative position between them. This relative position relation and the height difference are used to detect the curb direction as the local road direction. Finally, the occlusions are reconstructed using two different methods: (1) a cell-based linear interpolation and (2) a point-based mathematical morphology. The methodology is tested by original scanned data and multi-temporal evaluation data captured from a residential area in Delft, the Netherlands with vehicle-mounted LiDAR sensors. The result shows that all occlusions cause by vehicles are successfully detected and the curb (road) direction is correctly extracted in most of the occluded areas. Both reconstructed results can visually integrate the original scanned data and recover the curb structure. The reconstruction errors of the linear interpolation method are 0.045 m in the z-axis direction and 0.051 m in total and the reconstruction errors of mathematical morphology are 0.048 m in the z-axis direction and 0.052 m in total. • Ground boundaries are detected with an α-shape based on prior knowledge of vehicle. • Occlusions and vehicles are matched with the relative positions between them. • The local road direction is accurately estimated by detecting the curb direction. • A fast linear interpolation method is designed to reconstruct occlusions. • A point-based mathematical morphology method is proposed for reconstructions. [ABSTRACT FROM AUTHOR]

• Tree diversity is positively related to canopy space occupation in manged forests. • The definition of the canopy space and tree size largely affects diversity-canopy space occupation relationships. • Diversity effects are strongest by considering the upper canopy space. • Novel conceptual framework to assess spatial crown complementarity. A more efficient use of limited canopy space and, thus, a higher canopy space occupation (CSO) in forests can result in an increased absorption of photosynthetically active radiation, which in turn can promote productivity. Although there is some evidence for a positive relationship between tree diversity (TD) and CSO, the generality of this hypothesis is still under debate. Here, we propose a conceptual framework that accounts for both the spatial complexity of canopy space and size-dependent interspecific tree interactions and tested it using mobile laser scanning data across larger spatial scales. We assessed the CSO at high resolution with two diversity indices, tree species richness (TSR) and the effective number of species (ENS) along a TSR gradient ranging from monocultures to 8-species mixtures in a mature and structurally complex mixed-species temperate forest. We found that the direction and strength of the TD-CSO relationship largely depended on the way how canopy space is defined and which tree size classes are considered to calculate TSR. Using an broad deliniation of canopy space no significant relationship between TD and CSO was evident. In contrast, when considering only the upper canopy space, a significant effect of TSR on CSO emerged. Importantly, the direction of this relationship was critically dependent on the tree size threshold underlying the TSR determination. For all trees with a diameter at breast height > 7 cm, we observed a significant negative relationship, while the opposite was the case when considering only large-sized trees. Our novel conceptual framework demonstrates that accurate estimation of canopy space complexity and tree size dependence is key to better understanding the processes underlying CSO. However, further information on diversity-canopy space occupation relationships from studies in different forests and forest types is needed. [ABSTRACT FROM AUTHOR]

Various field-based studies of meander development have been realised, yet fluviomorphological processes and their spatial patterns over bends are not fully understood. In this paper, the processes of three natural meander bends during a flood event are dissected using a unique study approach, which combines traditional field surveying, close-range remote sensing, and computational modelling. The combined study approach improves the spatial and temporal resolutions of the data compared to traditional survey methods, although the bathymetric data collection needs improvement. The analyses indicate that the influence of a flood on the morphological changes of point bars depends on the stage of the bend's development. The point bars may also experience lateral growth during low discharge. However, the point bar is too large a unit to find simplified relationships between the discharge and morphological changes. Instead, one should concentrate on different parts of the bend and the point bar. In general, the longer the inundation period of one part, the more probable that the area will experience net deposition. The magnitude of the sediment transport and the net morphological change are not interconnected: Areas of net erosion and deposition can experience either high or low sediment transport rates during a flood. The results are compared with earlier theories of meander development, which are also assessed. [ABSTRACT FROM AUTHOR]

Line segment detection in images is already a well-investigated topic, although it has received considerably less attention in 3D point clouds. Benefiting from current LiDAR devices, large-scale point clouds are becoming increasingly common. Most human-made objects have flat surfaces. Line segments that occur where pairs of planes intersect give important information regarding the geometric content of point clouds, which is especially useful for automatic building reconstruction and segmentation. This paper proposes a novel method that is capable of accurately extracting plane intersection line segments from large-scale raw scan points. The 3D line-support region, namely, a point set near a straight linear structure, is extracted simultaneously. The 3D line-support region is fitted by our Line-Segment-Half-Planes (LSHP) structure, which provides a geometric constraint for a line segment, making the line segment more reliable and accurate. We demonstrate our method on the point clouds of large-scale, complex, real-world scenes acquired by LiDAR devices. We also demonstrate the application of 3D line-support regions and their LSHP structures on urban scene abstraction. [ABSTRACT FROM AUTHOR]

Point clouds collected in urban scenes contain a huge number of points (e.g., billions), numerous objects with significant size variability, complex and incomplete structures, and variable point densities, raising great challenges for the automated extraction of urban objects in the field of photogrammetry, computer vision, and robotics. This paper addresses these challenges by proposing an automated method to extract urban objects robustly and efficiently. The proposed method generates multi-scale supervoxels from 3D point clouds using the point attributes (e.g., colors, intensities) and spatial distances between points, and then segments the supervoxels rather than individual points by combining graph based segmentation with multiple cues (e.g., principal direction, colors) of the supervoxels. The proposed method defines a set of rules for merging segments into meaningful units according to types of urban objects and forms the semantic knowledge of urban objects for the classification of objects. Finally, the proposed method extracts and classifies urban objects in a hierarchical order ranked by the saliency of the segments. Experiments show that the proposed method is efficient and robust for extracting buildings, streetlamps, trees, telegraph poles, traffic signs, cars, and enclosures from mobile laser scanning (MLS) point clouds, with an overall accuracy of 92.3%. [ABSTRACT FROM AUTHOR]

Road markings are used to provide guidance and instruction to road users for safe and comfortable driving. Enabling rapid, cost-effective and comprehensive approaches to the maintenance of route networks can be greatly improved with detailed information about location, dimension and condition of road markings. Mobile Laser Scanning (MLS) systems provide new opportunities in terms of collecting and processing this information. Laser scanning systems enable multiple attributes of the illuminated target to be recorded including intensity data. The recorded intensity data can be used to distinguish the road markings from other road surface elements due to their higher retro-reflective property. In this paper, we present an automated algorithm for extracting road markings from MLS data. We describe a robust and automated way of applying a range dependent thresholding function to the intensity values to extract road markings. We make novel use of binary morphological operations and generic knowledge of the dimensions of road markings to complete their shapes and remove other road surface elements introduced through the use of thresholding. We present a detailed analysis of the most applicable values required for the input parameters involved in our algorithm. We tested our algorithm on different road sections consisting of multiple distinct types of road markings. The successful extraction of these road markings demonstrates the effectiveness of our algorithm. [ABSTRACT FROM AUTHOR]

Highlights: [•] We develop an automatic method for the segmentation of tunnel luminaires. [•] We use colored 3D point clouds acquired with a mobile terrestrial LiDAR system. [•] A four-stage algorithm based on color and geometrical features is proposed. [•] This automated method increases productivity of tunnel inspection works. [ABSTRACT FROM AUTHOR]

Although a scaffold is an essential structure in the construction industry, it may also be a dangerous factor that causes fatalities. However, the process of monitoring the scaffold is labor-intensive because it is conducted by the subjective observation of safety managers. To address this issue, we propose an automatic scaffold 3D reconstruction method using 3D point cloud data acquired using a robot dog. The method consists of three steps: 1) data acquisition of scaffold point clouds through a robot dog scanning system, 2) deep learning-based 3D semantic segmentation, and 3) automatic formation of a 3D CAD model. We created 15 robot dog datasets for training the segmentation model. The proposed method was tested at a different site where a scaffold with a representative structure was attached to the wall. The proposed method demonstrated an excellent performance of point cloud segmentation with a 90.84% F1 score. • The proposed method reconstructs a scaffold using a robot dog scanning system. • A robot dog efficiently acquired 3D point cloud of scaffold in various positions. • The method is based on the 3D semantic segmentation model with transfer learning. • In robot dog datasets, F1 score was 90.84%. [ABSTRACT FROM AUTHOR]

Abstract: In this paper we present an approach for automatic analysis of urban accessibility using 3D point clouds. Our approach is based on range images and it consists in two main steps: urban objects segmentation and curbs detection. Both of them are required for accessibility diagnosis and itinerary planning. Our method automatically segments facades and urban objects using two hypotheses: facades are the highest vertical structures in the scene and objects are bumps on the ground on the range image. The segmentation result is used to build an urban obstacle map. After that, the gradient is computed on the ground range image. Curb candidates are selected using height and geodesic features. Then, nearby curbs are reconnected using Bézier curves. Finally, accessibility is defined based on geometrical features and accessibility standards. Our methodology is tested on two MLS databases from Paris (France) and Enschede (The Netherlands). Our experiments show that our method has good detection rates, is fast and presents few false alarms. Our method outperforms other works reported in the literature on the same databases. [Copyright &y& Elsevier]

Abstract: Segmentation of mobile laser point clouds of urban scenes into objects is an important step for post-processing (e.g., interpretation) of point clouds. Point clouds of urban scenes contain numerous objects with significant size variability, complex and incomplete structures, and holes or variable point densities, raising great challenges for the segmentation of mobile laser point clouds. This paper addresses these challenges by proposing a shape-based segmentation method. The proposed method first calculates the optimal neighborhood size of each point to derive the geometric features associated with it, and then classifies the point clouds according to geometric features using support vector machines (SVMs). Second, a set of rules are defined to segment the classified point clouds, and a similarity criterion for segments is proposed to overcome over-segmentation. Finally, the segmentation output is merged based on topological connectivity into a meaningful geometrical abstraction. The proposed method has been tested on point clouds of two urban scenes obtained by different mobile laser scanners. The results show that the proposed method segments large-scale mobile laser point clouds with good accuracy and computationally effective time cost, and that it segments pole-like objects particularly well. [Copyright &y& Elsevier]

Abstract: Accurate 3D road information is important for applications such as road maintenance and virtual 3D modeling. Mobile laser scanning (MLS) is an efficient technique for capturing dense point clouds that can be used to construct detailed road models for large areas. This paper presents a method for extracting and delineating roads from large-scale MLS point clouds. The proposed method partitions MLS point clouds into a set of consecutive “scanning lines”, which each consists of a road cross section. A moving window operator is used to filter out non-ground points line by line, and curb points are detected based on curb patterns. The detected curb points are tracked and refined so that they are both globally consistent and locally similar. To evaluate the validity of the proposed method, experiments were conducted using two types of street-scene point clouds captured by Optech’s Lynx Mobile Mapper System. The completeness, correctness, and quality of the extracted roads are over 94.42%, 91.13%, and 91.3%, respectively, which proves the proposed method is a promising solution for extracting 3D roads from MLS point clouds. [Copyright &y& Elsevier]

The development of the IFC 4.3 candidate standard sets the pathway for interoperability amongst different infrastructures that have been harmonized under the buildingSMART project proposals. Furthermore, the use of laser scanning as a tool for the monitoring of infrastructures has also been rising. This paper showcases a combination of BIM and remote sensing for infrastructure in a road scenario. A semi-automatic approach is presented for the IFC 4.1 compliant modelling of traffic signs and guardrails, along with the inclusion of some semantics via property sets. The input data for the IFC modelling procedure is obtained from the automatic processing of a 3D point cloud, that extracts road alignment, traffic signs, and guardrails, and from standardized dimensions for certain assets. The point cloud processing is validated for the guardrail identification and the IFC modelling is described alongside an UML, exemplifying the possibility of modelling the as-is state of the road infrastructure. • IFC-compliant model generation of guardrails and traffic signs • IFC model of road asset containing target elements and additional semantics • Processing of 3D points clouds to extract guardrails and traffic signs • Guardrail extraction obtains F-score of 98% on the modelled point cloud [ABSTRACT FROM AUTHOR]

The amount of sight distance available to drivers is an extremely critical design element with significant impacts on the safe operation of highways. If the available sight distance on a roadway is limited, this reduces the ability of drivers to safely carry out certain maneuvers such as stopping or passing. Obstructions to sight distance are often caused by a range of different factors. This includes roadside vegetation overgrowing into the road's right of way and obstructions related to the road's alignment. To ensure that sufficient sight distance is available throughout a road's service life, transportation agencies are required to conduct regular long site visits to identify the existence of any obstructions while identifying the types of those obstructions. Although many studies have attempted quantifying available sight distance in the past, not much work has been done to automatically identify and study the types of obstructions causing sight distance limitations. This information is critical to entities looking to decide on the appropriate construction activity required to alleviate sight distance limitations. To fill in this gap, this paper proposes a fully automated novel method to assess sight distance on highways scanned using mobile LiDAR technology, while also proposing a method for the automated classification of the type of obstruction detected. The algorithm is tested on multiple roadway segments in Alberta, Canada and was successful in identifying limitations in sight distance and classifying obstructions into different types of objects. • Automated LiDAR based assessment of visibility on highways. • An algorithm for automated detection of sight distance obstructions is proposed. • The algorithm also automatically classifies obstructions into different objects. • High accuracy achieved after testing on multiple highway segments in Alberta. • Obstructions due to road, signs, vegetation...etc. are detected and classified. [ABSTRACT FROM AUTHOR]

During an emergency inside large buildings such as hospitals and shopping malls, the availability of up-to-date information is critical. One common source of information is the 2D layout of buildings and emergency exits. For most buildings, this information is represented as tangled floor plans, which in most cases are outdated. One solution to update the data of buildings after each renovation is to recreate 3D models of buildings in a quick and automatic approach. These 3D models provide proactively crucial building information in a digital format for first responders to be used in emergency cases. Thanks to advances in remote sensing, laser scanners can be used to generate an accurate spatial representation of buildings quickly. However, such devices provide point clouds, which are unstructured data. In this paper, we introduce a complete workflow that allows to generate 3D models from point clouds of buildings and extract fine-grained indoor navigation networks from those models, to support advanced path planning for disaster management and navigation of different types of agents. The process extracts structural elements of buildings such as walls, slabs, ceiling and openings, and reconstruct their volumetric shapes. Additionally, the furnishing elements in the input point clouds are identified and reconstructed as the obstacles. Stairs are also reconstructed to allow multistory navigation path planning. Our algorithm is fully 3D and can handle vertical and slanted structures. We test it on several real datasets, compared it to the state-of-the-art approaches and provide a process to check the consistency of the reconstruction, which allows in return to further improve its result. • A method is presented for indoor 3D reconstruction from point clouds of multi-story buildings. • A room reconstruction method is developed based on space-enclosure. • Modeling of furniture, obstacles, doors and stairs is included for optimal path finding. • The flexible space subdivision (FSS) provides a fine-grained navigation network in 2D and 3D. • Several heuristic control rules verify the consistency of the final 3D model. [ABSTRACT FROM AUTHOR]

The application of Deep Learning techniques to point clouds for urban object classification is limited by the large number of samples needed. Acquiring and tagging point clouds is more expensive and tedious labour than its image equivalent process. Point cloud online datasets contain few samples for Deep Learning or not always the desired classes This work focuses on minimizing the use of point cloud samples for neural network training in urban object classification. The method proposed is based on the conversion of point clouds to images (pc-images) because it enables: the use of Convolutional Neural Networks, the generation of several samples (images) per object (point clouds) by means of multi-view, and the combination of pc-images with images from online datasets (ImageNet and Google Images). The study is conducted with ten classes of objects extracted from two street point clouds from two different cities. The network selected for the job is the InceptionV3. The training set consists of 5000 online images with a variable percentage (0% to 10%) of pc-images. The validation and testing sets are composed exclusively of pc-images. Although the network trained only with online images reached 47% accuracy, the inclusion of a small percentage of pc-images in the training set improves the classification to 99.5% accuracy with 6% pc-images. The network is also applied at IQmulus & TerraMobilita Contest dataset and it allows the correct classification of elements with few samples. • Automatic point cloud urban object classification • Training set mainly formed by images obtained from online datasets • Conversion of point clouds to images to take advantage of Convolutional Neural Networks • 99.5% accuracy with only 6% images obtained from point clouds in the training set [ABSTRACT FROM AUTHOR]

Accurate data collected from on-going construction projects assists project field engineers with tracking the progress of construction work. The comparison of as-planned schedules against the as-built status of construction enables the involved parties to determine project performance. Periodic monitoring of project performance enables timely identification of discrepancies between the schedule baseline and the project schedule. The rapid identification of discrepancies allows necessary measures to be taken to minimize the impact of a delay on the construction workflow. In transportation projects, the traditional means of acquiring as-built data from a construction site prevents involved parties from receiving the required information from the site in a timely manner. The delay in the communication of information ultimately causes subsequent delays in implementing necessary courses of action targeted toward improving the workflow. Therefore, this project develops a technology-supplemented progress monitoring approach. The approach implements as-built data with a high level of detail is obtained from a jobsite within a reasonable period of time using lidar technology while ensuring the safety of the data collector. The proposed framework utilizes point cloud data obtained using mobile lidar technology and 4D design models to identify deviations of the performed work from the planned work. The framework was tested using as-built data acquired from an on-going bridge construction project, and Percentage of Completion (POC) for the as-built bridge elements are calculated and compared with the as-planned POC. The differences between these two POC values for every element, on a particular scan date, are used for assessing the performance of the proposed framework. The obtained POC difference values ranged from −7% to 6% for most elements, which shows that the developed framework enables tracking the completion of individual bridge elements in an accurate and efficient manner. • 3D design models, construction schedule and mobile lidar data were used to report project progress information. • POC for the as-built bridge elements are calculated and compared with the as-planned POC. • Individual bridge elements can be monitored accurately and efficiently using the proposed framework. • The manual processing time for each epoch is approximately 1 h and 15 min for the case study. [ABSTRACT FROM AUTHOR]

Shield subway tunnels are critical urban infrastructure and precise, rapid deformation detection method is required for the maintenance of these tunnels. A tunnel deformation detection system called Railway Mobile Measurement System (RMMS) was developed based on mobile laser scanning technique. Herein, detailed analysis is given to prove that profiles from the laser scanner can be seen as cross-sections of the tunnel and that the corresponding error can be ignored. A wavelet filtering algorithm was developed to filter the point cloud of tunnel ancillary facilities in the cross-section points. A subsequent ellipse fitting and deformation detection method is proposed. A cross-section location method for subway tunnels is also presented based on the greyscale image converted from the tunnel's point cloud. A tunnel ring segments seams detection method is proposed to detect tunnel ring segments dislocation. The accuracy of the system was verified in a subway tunnel in Wuhan, China with a total station. The accuracy of cross-section measurement was 1.5 mm, the repeatability of the overall and local deformation measurement was 0.5 and 0.9 mm, respectively. This system completely fulfils the accuracy requirement of railway tunnels deformation detection and its efficiency is significantly higher than that of the traditional method. • The error of directly obtained laser scanning cross-section points of subway tunnels is analyzed in detail. • A wavelet filtering algorithm is proposed to filter point cloud of tunnel ancillary facilities. • Tunnel's overall and local deformation is detected based on ellipse fitting of cross-section points. • A tunnel ring segments dislocation detection method based on template matching and segments seams detection is proposed. [ABSTRACT FROM AUTHOR]

Power line mapping using remote sensing can automate the traditionally labor-intensive power line corridor inspection. Land-based mobile laser scanning (MLS) can be a good choice for the power line mapping if an aerial inspection is impossible, too costly or slow, unsafe, prohibited by regulations, or if more detailed information on the power line corridor is needed. The mapping of the power lines using MLS was studied in a rural environment outside the road network for the first time. An automatic power line extraction algorithm was developed. The algorithm first found power line candidate points based on the shape and orientation of the local neighborhood of a point using principal component analysis. Power lines were retrieved from the candidates using random sample consensus (Ransac) and a new power line labeling method, which takes into account the three-dimensional shape of the power lines. The new labeling method was able to find the power lines and remove false detections, which were found, for example, from the forest. The algorithm was tested in forested and open field (arable land) areas, outside the road environment using two different platforms of MLS, namely, personal backpack and all-terrain vehicle. The recall and precision of the power line extraction were 93.3% and 93.6%, respectively, using 10 cm as a distance criterion for a successful detection. Drifting of the positioning solution of the scanner was the largest error source, being the (contributory) cause for 60–70% of the errors. The platform did not have a significant effect on the power line extraction accuracy. The accuracy was higher in the open field compared to the forest, because the one-dimensional point density along the power line was inhomogeneous and GNSS (global navigation satellite system) signal was weak in the forest. The results suggest that the power lines can be mapped accurately enough for inspection purposes using MLS in a rural environment outside the road network. • Power lines are mapped automatically from mobile laser scanning (MLS) point clouds. • MLS complements aerial power line inspection reducing costs and increasing quality. • Accuracy of automatic power line mapping using all-terrain MLS is higher than 93%. • Both all-terrain MLS platforms, ATV and personal backpack, provide similar accuracy. Accuracy is higher in the open field compared to forest because of GNSS outages. [ABSTRACT FROM AUTHOR]

Abstract: Road safety inspection is currently carried out by time-consuming visual inspection. The latest mobile mapping systems provide an efficient technique for acquiring very dense point clouds along road corridors, so that automated procedures for recognizing and extracting structures can be developed. This paper presents a framework for structure recognition from mobile laser scanned point clouds. It starts with an initial rough classification into three larger categories: ground surface, objects on ground, and objects off ground. Based on a collection of characteristics of point cloud segments like size, shape, orientation and topological relationships, the objects on ground are assigned to more detailed classes such as traffic signs, trees, building walls and barriers. Two mobile laser scanning data sets acquired by different systems are tested with the recognition methods. Performance analyses of the test results are provided to demonstrate the applicability and limits of the methods. While poles are recognized for up to 86%, classification into further categories requires further work and integration with imagery. [Copyright &y& Elsevier]