Search

The incorporation of detailed textures in 3D city models is crucial for enhancing their realism, as it adds depth and authenticity to the visual representation, thereby closely mimicking the surfaces and materials found in actual urban environments. Existing 3D city models can be enriched with energy-related roof and façade details, such as the material type (such as windows, green façades, bricks) and sunlight reflectance which can be derived from texture information. However, a common limitation of these models is their lack of very high resolution textures, which which reduces their realism and detail. Manually mapping textures onto each surface of a building is an exceptionally time-consuming and labor-intensive process, making it unfeasible for large-scale applications involving thousands of buildings. Therefore, an automated method is essential for texture mapping of 3D city models from aerial imagery. In this paper, we present CityJSON texture mapper – a python-based software tool for automated texture mapping of CityJSON-based 3D city models from oblique and nadir aerial imagery. Experimental results demonstrate the effectiveness of our approach in generating high-quality textured 3D city models, showcasing the potential for broader applications in geospatial analysis and decision-making. This research contributes to the ongoing efforts in enhancing the realism and usability of CityJSON-based 3D city models by enhancing them with their real textures from oblique aerial imagery. Texture mapped model can be explored at https://bit.ly/textured3dbag.

Asset management systems are beneficial for maintaining building infrastructure and can be used to keep up-to-date records of relevant safety assets, such as smoke detectors, exit signs, and fire extinguishers. Existing methods for locating and identifying these assets in buildings primarily rely on surveys and images, which only provide 2D locations and can be tedious for large-scale structures. Indoor point clouds, which can be captured quickly for buildings using mobile scanning techniques, can provide us with 3D asset locations. In this paper, we study the feasibility of using 3D point clouds of buildings combined with deep learning techniques to identify safety-related assets, particularly small-sized assets like fire switches and exit signs. We adopt the state-of-the-art Deep Learning network, Kernel Point-Fully Convolutional Network (KP-FCNN), to identify these assets through semantic segmentation. Using the obtained results, we create a 3D-geometry model of the building with assets pinpointed, providing scene semantics and delivering more value. Our method is tested using three different point cloud datasets obtained from a depth camera, a mobile laser scanner, and an iPhone lidar sensor.

Geometric errors in LoD2 building models can be caused by the modeling algorithm but are often related to the quality of input data. One approach to tackling the modeling errors caused by the quality of input data is to collect additional data with a UAV and remodel the buildings. However, no flight planning approach exists specifically designed for efficient data recollection for model improvement. In this paper, we propose an innovative flight planning approach for this purpose. Contrary to the conventional method that recollects the data covering the entire building roof, our approach only collects the data over the erroneous region and uses it to improve the erroneous model part later. Our algorithm utilizes the existing LiDAR survey data to automatically detect model errors and design the camera networks by considering the roof geometry. We optimize the trajectory that connects the viewpoints with a genetic algorithm and develops an obstacle avoidance function with ray-casting to ensure a collision-free path. The proposed flight plan is implemented in a real-world scene. Our result shows an improved point cloud created through dense image matching with the collected UAV image data. The generated point cloud is successfully used for creating partial building models for improving the original models.

Favelas are the most common type of informal settlements found in Brazil. The Housing Secretariat, City Hall, Sao Paulo, has conducted surveys using Unmanned Aerial Vehicles (UAVs) for the favelas to facilitate the slum upgrading projects and has taken the initiative to create a digital twin of the slum areas. This study illustrates the feasibility of developing a methodological workflow to create a digital twin by automatic 3D building reconstruction in slums from the UAV point cloud and the 2D building footprints with a continuous link for updating the building semantic information. This study focuses on facilitating data integration and updating the semantic information into the 3D model to provide additional information about the individual buildings in the slums. The assessments concluded that the proposed workflow is suitable for creating digital twins for slums based on the UAV and 2D cadastral data. However, the 3D slum model had a few limitations, which are discussed in this paper.

Precise and accurate three-dimensional geospatial data has become increasingly available thanks to advances in both Terrestrial Laser Scanning (TLS) and Structure-from-Motion Photogrammetry (SfM). These tools provide valuable information for mapping geomorphological features and detect surface changes in mountainous environments. The exploitation of 3D point-clouds has been proven tremendously useful in the field of geosciences. It remains, however, controversial whether cost efficient photogrammetry can provide as accurate and reliable geospatial information as the significantly more expensive laser scanning or not. In this study, a rockfall case site in the territory of Obergurgl, Austria, is investigated in order to provide answers to the above question in a complex environment. The analysis includes different terrestrial photogrammetry configurations aiming to comprehensively define the strengths and limitations of terrestrial photogrammetry over TLS. The latter constitutes an optimized methodology that provides guidelines for costly future assessments as part of the site investigation phase in geohazard management. There are no doubts that compared to traditional and conventional surveying methods TLS and Photogrammetry both offer products much faster and with a much higher data density. In the current study, we show that when photogrammetry is applied following a well-defined optimized strategy, it can be potentially an adequate alternative to more costly TLS datasets for mass movement assessment and monitoring purposes.

The 3rd edition of the international summer school “Close-range Sensing Techniques in Alpine terrain” took place in Obergurgl, Austria, in June 2019. This article reports on results from the training and seminar activities and the outcome of student questionnaire survey. Comparison between the recent edition and the past edition in 2017 shows no significant differences on the level of satisfaction on organizational and training aspects. Gender balance was present both in candidates and in the outcome of selections. Selection was based on past research activities and on topic relevance. The majority of trainees were therefore doctoral candidates and postdoctoral researchers, but also motivated master students participated. The training took place through keynotes, lectures, seminars, in the field with hands-on surveys followed by data analysis in the lab, and teamwork for preparing a final team presentation over different assignments.

Three-dimensional (3D) building model is gaining more scientific attention in recent times due to its application in various fields such as vehicle autonomous navigation, urban planning, heritage building documentation, gaming visualisation and tourism. The quality of the Level of Detail (LoD) of building models relies on the high-resolution data sets obtained for the building. As an alternative to laser scanners, Unmanned Aerial Vehicles (UAV) are efficient in collecting good quality images and generate reliable LoD3 of buildings (i.e. to model both roof and facades of a building) at comparatively lower cost and time. However, the complete collection of images on building facades is usually performed by manual flights along the different façade to assure a homogenous image coverage with the same resolution on each element: no offline autonomous procedure to define the main façade planes and acquire complete image sets independent of UAV platform have been developed yet. This paper proposes a novel methodology to generate the flight plan in correspondence of building facades. The Digital Surface Model (DSM) obtained from an initial nadir flight is used as an input to identify the target building and plan the image acquisition around it. The optimised flight plan ensures complete coverage of the building with a minimum number of images. The coordinates as well as the attitude of each planned image can be finally loaded on a UAV to perform the flight. In order to validate the proposed methodology, some tests performed on synthetic buildings of growing complexity and very different shapes are presented.

In this paper, a method is presented to improve the MLS platform’s trajectory for GNSS denied areas. The method comprises two major steps. The first step is based on a 2D image registration technique described in our previous publication. Internally, this registration technique first performs aerial to aerial image matching, this issues correspondences which enable to compute the 3D tie points by multiview triangulation. Similarly, it registers the rasterized Mobile Laser Scanning Point Cloud (MLSPC) patches with the multiple related aerial image patches. The later registration provides the correspondence between the aerial to aerial tie points and the MLSPC’s 3D points. In the second step, which is described in this paper, a procedure utilizes three kinds of observations to improve the MLS platform’s trajectory. The first type of observation is the set of 3D tie points computed automatically in the previous step (and are already available), the second type of observation is based on IMU readings and the third type of observation is soft-constraint over related pose parameters. In this situation, the 3D tie points are considered accurate and precise observations, since they provide both locally and globally strict constraints, whereas the IMU observations and soft-constraints only provide locally precise constraints. For 6DOF trajectory representation, first, the pose [R, t] parameters are converted to 6 B-spline functions over time. Then for the trajectory adjustment, the coefficients of B-splines are updated from the established observations. We tested our method on an MLS data set acquired at a test area in Rotterdam, and verified the trajectory improvement by evaluation with independently and manually measured GCPs. After the adjustment, the trajectory has achieved the accuracy of RMSE X = 9 cm, Y = 14 cm and Z = 14 cm. Analysing the error in the updated trajectory suggests that our procedure is effective at adjusting the 6DOF trajectory and to regenerate a reliable MLSPC product.

The 2nd international summer school “Close-range sensing techniques in Alpine terrain” was held in July 2017 in Obergurgl, Austria. Participants were trained in selected close-range sensing methods, such as photogrammetry, laser scanning and thermography. The program included keynotes, lectures and hands-on assignments combining field project planning, data acquisition, processing, quality assessment and interpretation. Close-range sensing was applied for different research questions of environmental monitoring in high mountain environments, such as geomorphologic process quantification, natural hazard management and vegetation mapping. The participants completed an online questionnaire evaluating the summer school, its content and organisation, which helps to improve future summer schools.

Pole-like road furniture detection received much attention due to its traffic functionality in recent years. In this paper, we develop a framework to detect pole-like road furniture from sparse mobile laser scanning data. The framework is carried out in four steps. The unorganised point cloud is first partitioned. Then above ground points are clustered and roughly classified after removing ground points. A slicing check in combination with cylinder masking is proposed to extract pole-like road furniture candidates. Pole-like road furniture are obtained after occlusion analysis in the last stage. The average completeness and correctness of pole-like road furniture in sparse and unevenly distributed mobile laser scanning data was above 0.83. It is comparable to the state of art in the field of pole-like road furniture detection in mobile laser scanning data of good quality and is potentially of practical use in the processing of point clouds collected by autonomous driving platforms.

The use of Indoor Mobile Laser Scanners (IMLS) for data collection in indoor environments has been increasing in the recent years. These systems, unlike Terrestrial Laser Scanners (TLS), collect data along a trajectory instead of at discrete scanner positions. In this research, we propose several methods to exploit the trajectories of IMLS systems for the interpretation of point clouds. By means of occlusion reasoning and use of trajectory as a set of scanner positions, we are capable of detecting openings in cluttered indoor environments. In order to provide information about both the partitioning of the space and the navigable space, we use the voxel concept for point clouds. Furthermore, to reconstruct walls, floor and ceiling we exploit the indoor topology and plane primitives. The results show that the trajectory is a valuable source of data for feature detection and understanding of indoor MLS point clouds.

Road furniture semantic labelling is vital for large scale mapping and autonomous driving systems. Much research has been investigated on road furniture interpretation in both 2D images and 3D point clouds. Precise interpretation of road furniture in mobile laser scanning data still remains unexplored. In this paper, a novel method is proposed to interpret road furniture based on their logical relations and functionalities. Our work represents the most detailed interpretation of road furniture in mobile laser scanning data. 93.3 % of poles are correctly extracted and all of them are correctly recognised. 94.3 % of street light heads are detected and 76.9 % of them are correctly identified. Despite errors arising from the recognition of other components, our framework provides a promising solution to automatically map road furniture at a detailed level in urban environments.

The use of Light Detection and Ranging (LiDAR) to study agricultural crop traits is becoming popular. Wheat plant traits such as crop height, biomass fractions and plant population are of interest to agronomists and biologists for the assessment of a genotype's performance in the environment. Among these performance indicators, plant population in the field is still widely estimated through manual counting which is a tedious and labour intensive task. The goal of this study is to explore the suitability of LiDAR observations to automate the counting process by the individual detection of wheat ears in the agricultural field. However, this is a challenging task owing to the random cropping pattern and noisy returns present in the point cloud. The goal is achieved by first segmenting the 3D point cloud followed by the classification of segments into ears and non-ears. In this study, two segmentation techniques: a) voxel-based segmentation and b) mean shift segmentation were adapted to suit the segmentation of plant point clouds. An ear classification strategy was developed to distinguish the ear segments from leaves and stems. Finally, the ears extracted by the automatic methods were compared with reference ear segments prepared by manual segmentation. Both the methods had an average detection rate of 85 %, aggregated over different flowering stages. The voxel-based approach performed well for late flowering stages (wheat crops aged 210 days or more) with a mean percentage accuracy of 94 % and takes less than 20 seconds to process 50,000 points with an average point density of 16  points/cm2. Meanwhile, the mean shift approach showed comparatively better counting accuracy of 95% for early flowering stage (crops aged below 225 days) and takes approximately 4 minutes to process 50,000 points.

Early career researchers such as PhD students are a main driving force of scientific research and are for a large part responsible for research innovation. They work on specialized topics within focused research groups that have a limited number of members, but might also have limited capacity in terms of lab equipment. This poses a serious challenge for educating such students as it is difficult to group a sufficient number of them to enable efficient knowledge transfer. To overcome this problem, the Innsbruck Summer School of Alpine Research 2015 on close-range sensing techniques in Alpine terrain was organized in Obergurgl, Austria, by an international team from several universities and research centres. Of the applicants a group of 40 early career researchers were selected with interest in about ten types of specialized surveying tools, i.e. laser scanners, a remotely piloted aircraft system, a thermal camera, a backpack mobile mapping system and different grade photogrammetric equipment. During the one-week summer school, students were grouped according to their personal preference to work with one such type of equipment under guidance of an expert lecturer. All students were required to capture and process field data on a mountain-related theme like landslides or rock glaciers. The work on the assignments lasted the whole week but was interspersed with lectures on selected topics by invited experts. The final task of the summer school participants was to present and defend their results to their peers, lecturers and other colleagues in a symposium-like setting. Here we present the framework and content of this summer school which brought together scientists from close-range sensing and environmental and geosciences.

Nowadays many cities and countries are creating their 3D building models for a better daily management and smarter decision making. The newly created 3D models are required to be consistent with existing 2D footprint maps. Thereby the 2D maps are usually combined with height data for the task of 3D reconstruction. Many buildings are often composed by parts that are discontinuous over height. Building parts can be reconstructed independently and combined into a complete building. Therefore, most of the state-of-the-art work on 3D building reconstruction first decomposes a footprint map into parts. However, those works usually change the footprint maps for easier partitioning and cannot detect building parts that are fully inside the footprint polygon. In order to solve those problems, we introduce two methodologies, one more dependent on height data, and the other one more dependent on footprints. We also experimentally evaluate the two methodologies and compare their advantages and disadvantages. The experiments use Airborne Laser Scanning (ALS) data and two vector maps, one with 1:10,000 scale and another one with 1:500 scale.

In mobile laser scanning systems, the platform’s position is measured by GNSS and IMU, which is often not reliable in urban areas. Consequently, derived Mobile Laser Scanning Point Cloud (MLSPC) lacks expected positioning reliability and accuracy. Many of the current solutions are either semi-automatic or unable to achieve pixel level accuracy. We propose an automatic feature extraction method which involves utilizing corresponding aerial images as a reference data set. The proposed method comprise three steps; image feature detection, description and matching between corresponding patches of nadir aerial and MLSPC ortho images. In the data pre-processing step the MLSPC is patch-wise cropped and converted to ortho images. Furthermore, each aerial image patch covering the area of the corresponding MLSPC patch is also cropped from the aerial image. For feature detection, we implemented an adaptive variant of Harris-operator to automatically detect corner feature points on the vertices of road markings. In feature description phase, we used the LATCH binary descriptor, which is robust to data from different sensors. For descriptor matching, we developed an outlier filtering technique, which exploits the arrangements of relative Euclidean-distances and angles between corresponding sets of feature points. We found that the positioning accuracy of the computed correspondence has achieved the pixel level accuracy, where the image resolution is 12cm. Furthermore, the developed approach is reliable when enough road markings are available in the data sets. We conclude that, in urban areas, the developed approach can reliably extract features necessary to improve the MLSPC accuracy to pixel level.

Automatic semantic interpretation of street furniture has become a popular topic in recent years. Current studies detect street furniture as connected components of points above the street level. Street furniture classification based on properties of such components suffers from large intra class variability of shapes and cannot deal with mixed classes like traffic signs attached to light poles. In this paper, we focus on the decomposition of point clouds of pole-like street furniture. A novel street furniture decomposition method is proposed, which consists of three steps: (i) acquirement of prior-knowledge, (ii) pole extraction, (iii) components separation. For the pole extraction, a novel global pole extraction approach is proposed to handle 3 different cases of street furniture. In the evaluation of results, which involves the decomposition of 27 different instances of street furniture, we demonstrate that our method decomposes mixed classes street furniture into poles and different components with respect to different functionalities.

Mobile Mapping (MM) is a technique to obtain geo-information using sensors mounted on a mobile platform or vehicle. The mobile platform’s position is provided by the integration of Global Navigation Satellite Systems (GNSS) and Inertial Navigation Systems (INS). However, especially in urban areas, building structures can obstruct a direct line-of-sight between the GNSS receiver and navigation satellites resulting in an erroneous position estimation. Therefore, derived MM data products, such as laser point clouds or images, lack the expected positioning reliability and accuracy. This issue has been addressed by many researchers, whose aim to mitigate these effects mainly concentrates on utilising tertiary reference data. However, current approaches do not consider errors in height, cannot achieve sub-decimetre accuracy and are often not designed to work in a fully automatic fashion. We propose an automatic pipeline to rectify MM data products by employing high resolution aerial nadir and oblique imagery as horizontal and vertical reference, respectively. By exploiting the MM platform’s defective, and therefore imprecise but approximate orientation parameters, accurate feature matching techniques can be realised as a pre-processing step to minimise the MM platform’s three-dimensional positioning error. Subsequently, identified correspondences serve as constraints for an orientation update, which is conducted by an estimation or adjustment technique. Since not all MM systems employ laser scanners and imaging sensors simultaneously, and each system and data demands different approaches, two independent workflows are developed in parallel. Still under development, both workflows will be presented and preliminary results will be shown. The workflows comprise of three steps; feature extraction, feature matching and the orientation update. In this paper, initial results of low-level image and point cloud feature extraction methods will be discussed as well as an outline of the project and its framework will be given.

The European FP7 project IQmulus yearly organizes several processing contests, where submissions are requested for novel algorithms for point cloud and other big geodata processing. This paper describes the set-up and execution of a contest having the purpose to evaluate state-of-the-art algorithms for Mobile Mapping System point clouds, in order to detect and identify (individual) trees. By the nature of MMS these are trees in the vicinity of the road network (rather than in forests). Therefore, part of the challenge is distinguishing between trees and other objects, such as buildings, street furniture, cars etc. Three submitted segmentation and classification algorithms are thus evaluated.

The Multi-View Stereo (MVS) technology has improved significantly in the last decade, providing a much denser and more accurate point cloud than before. The point cloud now becomes a valuable data for modelling the LOD2 buildings. However, it is still not accurate enough to replace the lidar point cloud. Its relative high level of noise prevents the accurate interpretation of roof faces, e.g. one planar roof face has uneven surface of points therefore is segmented into many parts. The derived roof topology graphs are quite erroneous and cannot be used to model the buildings using the current methods based on roof topology graphs. We propose a parameter-free algorithm to robustly and precisely derive roof structures and building models. The points connecting roof segments are searched and grouped as structure points and structure boundaries, accordingly presenting the roof corners and boundaries. Their geometries are computed by the plane equations of their attached roof segments. If data available, the algorithm guarantees complete building structures in noisy point clouds and meanwhile achieves global optimized models. Experiments show that, when comparing to the roof topology graph based methods, the novel algorithm achieves consistent quality for both lidar and photogrammetric point clouds. But the new method is fully automatic and is a good alternative for the model-driven method when the processing time is important.

This paper describes a method that aims to find all instances of a certain object in Mobile Laser Scanner (MLS) data. In a userassisted approach, a sample segment of an object is selected, and all similar objects are to be found. By selecting samples from multiple classes, a classification can be performed. Key assumption in this approach is that a one-to-one relationship exists between segments and objects. In this paper the focus is twofold: (1) to explain how to get proper segments, and (2) to describe how to find similar objects. Point attributes that help separating neighbouring objects are presented. These point attributes are used in an attributed connected component algorithm where segments are grown, based on proximity and attribute values. Per component, a feature vector is proposed that consists of two parts. The first is a height histogram, containing information on the height distribution of points within a component. The second contains size and shape information, based on the components’ bounding box. A simple correlation function is used to find similarities between samples, as selected by a user, and other components. Our approach is tested on a MLS dataset, containing over 300 objects in 13 classes. Detection accuracies heavily depend on the success of the segmentation, and the number of selected samples in combination with the variety of object types in the scene.

We present a method for detecting and modelling rails in mobile laser scanner data. The detection is based on the properties of the rail tracks and contact wires such as relative height, linearity and relative position with respect to other objects. Points classified as rail track are used in a 3D modelling algorithm. The modelling is done by first fitting a parametric model of a rail piece to the points along each track, and estimating the position and orientation parameters of each piece model. For each position and orientation parameter a smooth low-order Fourier curve is interpolated. Using all interpolated parameters a mesh model of the rail is reconstructed. The method is explained using two areas from a dataset acquired by a LYNX mobile mapping system in a mountainous area. Residuals between railway laser points and 3D models are in the range of 2 cm. It is concluded that a curve fitting algorithm is essential to reliably and accurately model the rail tracks by using the knowledge that railways are following a continuous and smooth path.

Due to the road safety problem is becoming more and more serious recent years, existing road safety assessment by using automatic method is necessary. Meanwhile, since the pole-like objects have large effect on road safety and are in high demand as facilities to be managed, the automatic pole-like objects extraction is becoming a hot issue. As a result, a robust, quick and automatic pole-like object detection algorithm in MLS data is proposed in this paper. Two datasets are tested to show performance of the proposed algorithm, it demonstrates that it is feasible to detect tree with an overall accuracy of over 92% and other pole-like object of 72% in dataset A and 82% of tree points and 75% of other pole points in dataset B.

The reconstruction of 3D building models has been extensively researched in the last decade. The model driven methods, which fit building parts to predefined building primitives and glue them together into complete buildings, have been proved to be powerful. This paper solves two open questions left in prior works. Firstly, the primitive building libraries defined in literature are limited to few types of building structures. Therefore the undefined types of buildings could not be well represented by the primitive libraries. Secondly, the roof topology graphs used for interpreting the roof structures are assumed to be correct in previous works. In many cases, however, the roof topology graphs are inevitably wrong because of outliers in input data and errors in their construction steps. In this paper we solve the two problems and compare the improvements with prior woks. We introduce a complete building library, which can present planer buildings with arbitrary structures, and present an automatic method to correct the erroneous roof topology graph. Experiments show our method is robust to reconstruct buildings in both resident and commercial areas.

Building change detection serves to investigate illegal buildings. Illegal built or removed structures, especially those concealed among gable roofs such as dormers, are difficult to track among potentially millions of buildings. Nevertheless, they can be efficiently located in changed areas. An approach is proposed to automatically detect and classify changes in buildings from two epochs of Airborne Laser Scanning Data. Both datasets are classified into water, ground, building, vegetation and undefined objects in advance. After generalization of a 3D surface separation map, we verify changes by making rules on the separation map. Changes belonging to buildings are then classified into roof, wall, dormers, vehicles, construction above roof and undefined objects. As the ALS data has accuracy in strip difference of lower than 5 cm within the same epoch and from different epochs, changes that are larger than 10 cm were detected. Building changes, which areas are larger than 4 m2, are identified as change. By inspection, nearly all changes are detected and approximately 80% changes are correctly classified.

Water run-off modelling applied within urban areas requires an appropriate detailed surface model represented by a raster height grid. Accurate simulations at this scale level have to take into account small but important water barriers and flow channels given by the large-scale map definitions of buildings, street infrastructure, and other terrain objects. Thus, these 3D features have to be rasterised such that each cell represents the height of the object class as good as possible given the cell size limitations. Small grid cells will result in realistic run-off modelling but with unacceptable computation times; larger grid cells with averaged height values will result in less realistic run-off modelling but fast computation times. This paper introduces a height grid generalisation approach in which the surface characteristics that most influence the water run-off flow are preserved. The first step is to create a detailed surface model (1:1.000), combining high-density laser data with a detailed topographic base map. The topographic map objects are triangulated to a set of TIN-objects by taking into account the semantics of the different map object classes. These TIN objects are then rasterised to two grids with a 0.5m cell-spacing: one grid for the object class labels and the other for the TIN-interpolated height values. The next step is to generalise both raster grids to a lower resolution using a procedure that considers the class label of each cell and that of its neighbours. The results of this approach are tested and validated by water run-off model runs for different cellspaced height grids at a pilot area in Amersfoort (the Netherlands). Two national datasets were used in this study: the large scale Topographic Base map (BGT, map scale 1:1.000), and the National height model of the Netherlands AHN2 (10 points per square meter on average). Comparison between the original AHN2 height grid and the semantically enriched and then generalised height grids shows that water barriers are better preserved with the new method. This research confirms the idea that topographical information, mainly the boundary locations and object classes, can enrich the height grid for this hydrological application.

We aim at efficiently classifying ALS data in urban areas by choosing an optimal combination of features and entities. Three kinds of entities are defined, namely, single points, planar segments and segments obtained by mean-shift segmentation. Various features are computed for these three entities. All derived features are assigned to different steps of our method. Our method is composed of a sequence of rule based classifications. After a rule based classification for planar segments and a context rule based classification for walls and roof elements 85% of the data are well classified. Errors mainly appear in the area where rules are difficult to define, such as vegetation close to walls and above roofs. To eliminate these errors, we first group all the points in these areas into segments using mean shift, and then search for segments with potentially misclassified points using a distance ratio. These mean shift segments are then re-classified using another rule based classification. The overall quality of our classification method reaches to 98.1%.

Indoor space subdivision is an important aspect of scene analysis that provides essential information for many applications, such as indoor navigation and evacuation route planning. Until now, most proposed scene understanding algorithms have been based on whole point clouds, which has led to complicated operations, high computational loads and low processing speed. This paper presents novel methods to efficiently extract the location of openings (e.g., doors and windows) and to subdivide space by analyzing scanlines. An opening detection method is demonstrated that analyses the local geometric regularity in scanlines to refine the extracted opening. Moreover, a space subdivision method based on the extracted openings and the scanning system trajectory is described. Finally, the opening detection and space subdivision results are saved as point cloud labels which will be used for further investigations. The method has been tested on a real dataset collected by ZEB-REVO. The experimental results validate the completeness and correctness of the proposed method for different indoor environment and scanning paths.

This study develops an integrated data-driven and model-driven approach (template matching) that clusters the urban railroad point clouds into three classes of rail track, contact cable, and catenary cable. The employed dataset covers 630 m of the Dutch urban railroad corridors in which there are four rail tracks, two contact cables, and two catenary cables. The dataset includes only geometrical information (three dimensional (3D) coordinates of the points) with no intensity data and no RGB data. The obtained results indicate that all objects of interest are successfully classified at the object level with no false positives and no false negatives. The results also show that an average 97.3% precision and an average 97.7% accuracy at the point cloud level are achieved. The high precision and high accuracy of the rail track classification (both greater than 96%) at the point cloud level stems from the great impact of the employed template matching method on excluding the false positives. The cables also achieve quite high average precision (96.8%) and accuracy (98.4%) due to their high sampling and isolated position in the railroad corridor., Competing Interests: The authors declare no conflict of interest.