Although point clouds are characterized as a type of unstructured data, timestamp attribute can structure point clouds into scanlines and shape them into a time signal. The present work studies the transformation of the street point cloud into a time signal based on the Z component for the semantic segmentation using Long Short-Term Memory (LSTM) networks. The experiment was conducted on the point cloud of a real case study. Several training sessions were performed changing the Level of Detail of the classification (coarse level with 3 classes and fine level with 11 classes), two levels of network depth and the use of weighting for the improvement of classes with low number of points. The results showed high accuracy, reaching at best 97.3% in the classification with 3 classes (ground, buildings, and objects) and 95.7% with 11 classes. The distribution of the success rates was not the same for all classes. The classes with the highest number of points obtained better results than the others. The application of weighting improved the classes with few points at the expense of the classes with more points. Increasing the number of hidden layers was shown as a preferable alternative to weighting. Given the high success rates and a behaviour of the LSTM consistent with other Neural Networks in point cloud processing, it is concluded that the LSTM is a feasible alternative for the semantic segmentation of point clouds transformed into time signals.

Traffic signs are one of the most relevant road assets for driving, as the safety of drivers depends to a great extent on their correct location. In this paper two methods are compared for the segmentation of the sign and the pole supporting it. Both methods are based on the morphological opening to identify the sign points, the first one directly employs the mathematical morphology directly applied to point clouds and the second one through point cloud rasterization into images. The comparison was conducted on twenty real traffic signs acquired with Mobile Laser Scanning obtaining point clouds from environments with signposts, traffic lights and lampposts. The results showed a correct segmentation of the signs, obtaining a F-score of 0.81 by the point-based method and a 0.75 by 2D image method. In particular, the point-based mathematical morphology proved to be more accurate in the segmentation of traffic sings installed on traffic lights and lampposts, avoiding over detection shown by the 2D image method.

Mathematical morphology is a technique recently applied directly for point cloud data. Its working principle is based on the removal and addition of points from an auxiliary point cloud that acts as a structuring element. However, in certain applications within a more complex process, these changes to the original data represent an unacceptable loss of information. The aim of this work is to provide a modification of the morphological opening to retain original points and attributes. The proposed amendment involved in the morphological opening: erosion followed by dilatation. In morphological erosion, the new eroded points are retained. In morphological dilation, the structuring element does not add its points directly, but uses the point positions to search through the previously eroded points and retrieve them for the dilated point cloud. The modification was tested on synthetic and real data, showing a correct performance at the morphological level, and preserving the precision of the original points and their attributes. Furthermore, the conservation is shown to be very relevant in two possible applications such as traffic sign segmentation and occluded edge detection.

Occlusions accompany serious problems that reduce the applicability of numerous algorithms. The aim of this work is to detect and characterize urban ground gaps based on occluding object. The point clouds for input have been acquired with Mobile Laser Scanning and have been previously segmented into ground, buildings and objects, which have been classified. The method generates various raster images according to segmented point cloud elements, and detects gaps within the ground based on their connectivity and the application of the hit-or-miss transform. The method has been tested in four real case studies in the cities of Vigo and Paris, and an accuracy of 99.6% has been obtained in occlusion detection and labelling. Cars caused 80.6% of the occlusions. Each car occluded an average ground area of 11.9 m2. The proposed method facilitates knowing the percentage of occluded ground, and if this would be reduced in successive multi-temporal acquisitions based on mobility characteristics of each object class.

Indoor furniture is of great relevance to building occupants in everyday life. Furniture occupies space in the building, gives comfort, establishes order in rooms and locates services and activities. Furniture is not always static; the rooms can be reorganized according to the needs. Keeping the building models up to date with the current furniture is key to work with indoor environments. Laser scanning technology can acquire indoor environments in a fast and precise way, and recent artificial intelligence techniques can classify correctly the objects that contain. The objective of this work is to study how to minimize the use of point cloud samples in Neural Network training, tedious to label, and replace them with images obtained from online sources. For this, point clouds are converted to images by means of rotations and projections. The conversion of a 3D vector data to a 2D raster allows the use of Convolutional Neural Networks, the achievement of several images for each acquired point cloud object and the combination with images obtained from online sources, such as Google Images. The images have been distributed among the validation and testing training sets following different percentages. The results show that, although point cloud images cannot be completely dispensed within the training set, only 10% of these achieve high accuracy in the classification.

Room segmentation is a matter of ongoing interesting for indoor navigation and reconstruction in robotics and AEC. While in robotics field, the problem room segmentation has been typically addressed on 2D floorplan, interest in enrichment 3D models providing more detailed representation of indoors has been growing in the AEC. Point clouds make available more realistic and update but room segmentation from point clouds is still a challenging topic. This work presents a method to carried out point cloud segmentation into rooms based on 3D mathematical morphological operations. First, the input point cloud is voxelized and indoor empty voxels are extracted by CropHull algorithm. Then, a morphological erosion is performed on the 3D image of indoor empty voxels in order to break connectivity between voxels belonging to adjacent rooms. Remaining voxels after erosion are clustered by a 3D connected components algorithm so that each room is individualized. Room morphology is retrieved by individual 3D morphological dilation on clustered voxels. Finally, unlabelled occupied voxels are classified according proximity to labelled empty voxels after dilation operation. The method was tested in two real cases and segmentation performance was evaluated with encouraging results.

Intercomparison among six terrestrial laser scanner systems focused on the measurement of small elements (< 0.5 m) is performed. Phase shift (PS) and time of flight (ToF) scanners are considered. Two standard artefacts containing threedimensional printing spheres and steps of variable height are used for the experiment. Results show errors between -4.5 and 3.5 mm in the measurement of distances between step planes. The most stable systems for measuring small elements seem the Leica C10, Faro Photon and Riegl LMS Z390i. The quality of the results is linked to the overall quality of the system rather than the specific technology used for range measurement (PS or ToF) which does not appear to be a determining factor.

Accessibility diagnosis of as-built urban environments is essential for path planning, especially in case of people with reduced mobility and it requires an in-depth knowledge of ground elements. In this paper, we present a new approach for automatically detect and classify urban ground elements from 3D point clouds. The methodology enables a high level of detail classification from the combination of geometric and topological information. The method starts by a planar segmentation followed by a refinement based on split and merge operations. Next, a feature analysis and a geometric decision tree are followed to classify regions in preliminary classes. Finally, adjacency is studied to verify and correct the preliminary classification based on a comparison with a topological graph library. The methodology is tested in four real complex case studies acquired with a Mobile Laser Scanner Device. In total, five classes are considered (roads, sidewalks, treads, risers and curbs). Results show a success rate of 97% in point classification, enough to analyse extensive urban areas from an accessibility point of view. The combination of topology and geometry improves a 10% to 20% the success rate obtained with only the use of geometry.

Visibility is a common measure to describe the spatial properties of an environment related to the spatial behaviour. Isovists represent the space that can be seen from one observation point, and they are used to analyse the existence of obstacles affecting or blocking intervisibility in an area. Although point clouds depict the as-built reality in a very detailed and accurate way, literature addressing the analysis of visibility in 3D, and more specifically the usage of point clouds to visibility analysis, is rather limited. In this paper, a methodology to evaluate visibility from point clouds in indoor environments is proposed, resulting in the creation of 3D isovists. Point cloud is firstly discretized in a voxel-based structure and voxels are labelled into ‘exterior’, ‘occupied’, ‘visible’ and ‘occluded’ based on an occupancy followed by a visibility analysis performed from a ray-tracing algorithm. 3D Isovists are created from the boundary of visible voxels from an observer position and considering as input parameters the visual angle, maximum line of sight, and eye gaze direction.

This work presents an approach to automatically detect structural floor elements such as steps or ramps in the immediate environment of buildings, elements that may affect the accessibility to buildings. The methodology is based on Mobile Laser Scanner (MLS) point cloud and trajectory information. First, the street is segmented in stretches along the trajectory of the MLS to work in regular spaces. Next, the lower region of each stretch (the ground zone) is selected as the ROI and normal, curvature and tilt are calculated for each point. With this information, points in the ROI are classified in horizontal, inclined or vertical. Points are refined and grouped in structural elements using raster process and connected components in different phases for each type of previously classified points. At last, the trajectory data is used to distinguish between road and sidewalks. Adjacency information is used to classify structural elements in steps, ramps, curbs and curb-ramps. The methodology is tested in a real case study, consisting of 100 m of an urban street. Ground elements are correctly classified in an acceptable computation time. Steps and ramps also are exported to GIS software to enrich building models from Open Street Map with information about accessible/inaccessible entrances and their locations.

The thermal analysis of buildings represents a key-step for reduction of energy consumption, also in the case of Cultural Heritage. Here the complexity of the constructions and the adopted materials might require special analysis and tailored solutions. Infrared Thermography (IRT) is an important non-destructive investigation technique that may aid in the thermal analysis of buildings. The paper reports the application of IRT on a listed building, belonging to the Cultural Heritage and to a residential one, as a demonstration that IRT is a suitable and convenient tool for analysing the existing buildings. The purposes of the analysis are the assessment of the damages and energy efficiency of the building envelope. Since in many cases the complex geometry of historic constructions may involve the thermal analysis, the integration of IRT and accurate 3D models were developed during the latest years. Here authors propose a solution based on the up-To-date photogrammetric solutions for purely image-based 3D modelling, including automatic image orientation/sensor calibration using Structure-from-Motion and dense matching. Thus, an almost fully automatic pipeline for the generation of accurate 3D models showing the temperatures on a building skin in a realistic manner is described, where the only manual task is given by the measurement of a few common points for co-registration of RGB and IR photogrammetric projects.

Updated and detailed indoor models are being increasingly demanded for various applications such as emergency management or navigational assistance. The consolidation of new portable and mobile acquisition systems has led to a higher availability of 3D point cloud data from indoors. In this work, we explore the combined use of point clouds and trajectories from SLAM-based laser scanner to automate the reconstruction of building indoors. The methodology starts by door detection, since doors represent transitions from one indoor space to other, which constitutes an initial approach about the global configuration of the point cloud into building rooms. For this purpose, the trajectory is used to create a vertical point cloud profile in which doors are detected as local minimum of vertical distances. As point cloud and trajectory are related by time stamp, this feature is used to subdivide the point cloud into subspaces according to the location of the doors. The correspondence between subspaces and building rooms is not unambiguous. One subspace always corresponds to one room, but one room is not necessarily depicted by just one subspace, for example, in case of a room containing several doors and in which the acquisition is performed in a discontinue way. The labelling problem is formulated as combinatorial approach solved as a minimum energy optimization. Once the point cloud is subdivided into building rooms, envelop (conformed by walls, ceilings and floors) is reconstructed for each space. The connectivity between spaces is included by adding the previously detected doors to the reconstructed model. The methodology is tested in a real case study.

3D models of indoor environments are essential for many application domains such as navigation guidance, emergency management and a range of indoor location-based services. The principal components defined in different BIM standards contain not only building elements, such as floors, walls and doors, but also navigable spaces and their topological relations, which are essential for path planning and navigation. We present an approach to automatically reconstruct topological relations between navigable spaces from point clouds. Three types of topological relations, namely containment, adjacency and connectivity of the spaces are modelled. The results of initial experiments demonstrate the potential of the method in supporting indoor navigation.

Automated generation of 3D indoor models from point cloud data has been a topic of intensive research in recent years. While results on various datasets have been reported in literature, a comparison of the performance of different methods has not been possible due to the lack of benchmark datasets and a common evaluation framework. The ISPRS benchmark on indoor modelling aims to address this issue by providing a public benchmark dataset and an evaluation framework for performance comparison of indoor modelling methods. In this paper, we present the benchmark dataset comprising several point clouds of indoor environments captured by different sensors. We also discuss the evaluation and comparison of indoor modelling methods based on manually created reference models and appropriate quality evaluation criteria. The benchmark dataset is available for download at: http://www2.isprs.org/commissions/comm4/wg5/benchmark-on-indoor-modelling.HTML.

Indoor localization has gained considerable attention over the past decade because of the emergence of numerous location-aware services. Research works have been proposed on solving this problem by using wireless networks. Nevertheless, there is still much room for improvement in the quality of the proposed classification models. In the last years, the emergence of Visible Light Communication (VLC) brings a brand new approach to high quality indoor positioning. Among its advantages, this new technology is immune to electromagnetic interference and has the advantage of having a smaller variance of received signal power compared to RF based technologies. In this paper, a performance analysis of seventeen machine leaning classifiers for indoor localization in VLC networks is carried out. The analysis is accomplished in terms of accuracy, average distance error, computational cost, training size, precision and recall measurements. Results show that most of classifiers harvest an accuracy above 90 %. The best tested classifier yielded a 99.0 % accuracy, with an average error distance of 0.3 centimetres.