Mobile Laser Scanning Datasets Research Articles

Ground Surface Recognition at Voxel Scale From Mobile Laser Scanning Data in Urban Environment

Ground filtering is an essential process for further classification of other features from a point cloud. This letter aims to develop an efficient method for the automatic recognition of a ground surface in urban environments from a point cloud acquired by mobile laser scanning (MLS). The MLS point cloud has large amounts of data. To decrease the calculation effort, the point clouds are first segmented into regular voxels using an octree structure. Then, voxel cloud connectivity segmentation (VCCS) is applied to generate supervoxels, which alleviate the issue of boundary cross between voxels and ground truth. Finally, several geometric features on the voxel scale are selected for a support vector machine (SVM) to label segments as a ground surface and nonground objects. For an objective comparison with other methods, the accuracy of the proposed method is evaluated by public MLS data sets. Experimental results show that all ground type objects can be well recognized with least susceptibility to scenic change.

Identification of trees and their trunks from mobile laser scanning data of roadway scenes

ABSTRACTTrees along the roads are important assets, which need continuous assessment and maintenance. The mobile laser scanning (MLS) has been adopted as mainstream mapping technique for three-dimensional data acquisition along the roads. In this study, an automated method was developed to identify trees and their trunks from MLS data. A bottom-up search in two stages is adopted in the cylinders, which are formed by partitioning of normalized MLS data. Tree trunk is identified first based on linearity and data distribution homogeneity along lower section of object clusters lying near to the respective cylinder’s base centre. Then, crown of tree is retrieved for respective identified trunk using compactness index for circular or near-circular cross section of crown and its axial symmetry about trunk axis. The object cluster composed of trunk and crown both are identified as tree. The proposed method was tested and validated on MLS data of two different roadway test sites that were acquired at different point spacing. The results reveal that the performance of proposed method in these two sites in terms of average completeness, correctness, and measure was 94.4%, 100%, and 97.1%, respectively. The correctness did not change in both sites and it was 100% and stable, which showed that none of the non-tree objects was falsely identified as tree and correctness in trees identification was independent of the test site complexity. The proposed method holds great potential for identifying trees from MLS data of various roadway site conditions, where shapes and sizes of trees in their 3D data get distorted due to occlusions, and partial overlap presents among objects. Furthermore, the proposed method was implemented in the graphics processing unit-based parallel computing framework and runtime was dramatically minimized on MLS datasets of two test sites.

Recovery of urban 3D road boundary via multi-source data

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.

Automatic extraction of accurate 3D tie points for trajectory adjustment of mobile laser scanners using aerial imagery

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.

Precise Registration of Laser Mapping Data by Planar Feature Extraction for Deformation Monitoring

Quantifying near-field displacements can help enable a better understanding of earthquake physics and hazards. To date, established remote sensing techniques have failed to recover subcentimeter-level near-field displacements at the scale and resolution required for shallow fault physical investigations. In this paper, methods are developed to rapidly extract planar parameters, using fast parallel approaches and an alternative registration approach is employed to automatically match the planes extracted from pairwise temporally spaced mobile laser scanning (MLS) and Airborne laser scanning (ALS) data sets along the Napa fault. The features extracted from two temporally spaced point clouds are then used to calculate rigid-body transformation parameters. The production of robust and accurate deformation maps requires the selection of appropriate planar feature extraction and feature mapping criteria. Rigorously propagated point accuracy estimates are employed to produce realistic estimated errors for the transformation parameters. Displacements of each aggregate study area are computed separately from left and right sides of the fault and compared to be within 3 mm of alinement array displacements. Local differential displacements show distinct patterns which, compared to alinement array measurements, were found to agree within the confidence bounds. The findings demonstrate the ability to accurately estimate near-field deformations from repeated MLS or ALS scans of earthquake-prone urban areas. ALS is also used in conjunction with the MLS data sets, illustrating the algorithm’s ability to accommodate different LiDAR collection modalities at subcentimeter-level accuracy. The automated planar extraction and registration is an important contribution to the study of near-field earthquake dynamics and can be used as input observations for future geological inversion models.

Structural segmentation and classification of mobile laser scanning point clouds with large variations in point density

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 F1-score of 72.70% of seven common classes, being higher than the best of previous results.

Development of deep learning architecture for automatic classification of outdoor mobile LiDAR data

ABSTRACTThis paper proposes a deep convolutional neural network (CNN) architecture for automatic classification of mobile laser scanning (MLS) data obtained for outdoor environment, which are characterized by noise, clutter, large size and larger quantum of information. The developed architecture introduces a look up table (LUT) based approach, which retains the geometry of the input MLS point cloud while rescaling. Further, with the voxelisation of the input MLS sample, the ambiguity of selecting one out of multiple point values within a voxel is resolved. The performance of the architecture is evaluated on MLS data of outdoor environment in two instances, first using tree and non-tree classes (non-tree class has objects like electric pole, wire, low vegetation, wall, house and ground) and then with tree and electric pole classes. Additional testing is carried out by mixing the outdoor MLS data of tree and electric pole classes with three classes of indoor objects, taken from Modelnet dataset, thereby assessing the architecture efficacy over an ensemble of three-dimensional (3D) datasets. Classification of tree and non-tree classes, followed by tree and electric pole classes from MLS samples result in total accuracies of 86.0%, 90.0% respectively and kappa values of 72.0%, 78.7% respectively. Moreover, for the combinations of MLS and Modelnet classes, the classification results are promising, reaching a total accuracy of 95.2% and kappa of 92.5%. The LUT based approach has shown better classification over the traditional rescaling approach for the MLS dataset, resulting in an enhancement up to 9.0% and 18.0% in total accuracy and kappa, respectively. With different varieties of tree, non-tree and electric pole samples, the proposed architecture has shown its potential for automatic classification of MLS data with high accuracy. This study further reveals that the accuracy of classification is improved by introducing more spatial features in the input layer. The accuracies produced in this work can be further improved with the availability of better hardware resources.

Automated detection and decomposition of railway tunnels from Mobile Laser Scanning Datasets

Abstract Since the mid-19th century, the railway network has occupied a crucial place at the heart of the world's transport systems. Its infrastructure is often situated in harsh environments where an extreme event, or even daily use, could lead to a catastrophic accident. This is one of the main reasons why inspecting these constructions is so important. Despite the advances in this field, the human component continues to be part of the final inspection process. In order to improve on this, this paper shows the use of laser scanning as a leading technology in automating the inspection of railway infrastructures. The proposed methodologies provide the essential processed and classified data needed for the structural health monitoring of the various assets related to railways. It is divided into three main parts, which pre-process the point cloud, divide the cloud into ground and non-ground points, and detect the elements present in each of these clouds. The methods are validated in three case studies, each containing different railway tunnels. The results demonstrate that laser scanning technology, together with customized processing tools, can provide data for further structural operations with no requirement for either training in geomatics or high-performance computers for the data processing. Significant results are obtained for the developed classification methods: the classification of the tunnel elements returns a global F-Score of between 71 and 99% in a point-by-point comparison. With regard to the labelled rails classification, a global F-Score of 100% is achieved for the analyzed datasets, and between 56 and 73% for the point-by-point classification.

Reduction Method for Mobile Laser Scanning Data

Mobile Laser Scanning (MLS) technology acquires a huge volume of data in a very short time. In many cases, it is reasonable to reduce the size of the dataset with eliminating points in such a way that the datasets, after reduction, meet specific optimization criteria. Various methods exist to decrease the size of point cloud, such as raw data reduction, Digital Terrain Model (DTM) generalization or generation of regular grid. These methods have been successfully applied on data captured from Airborne Laser Scanning (ALS) and Terrestrial Laser Scanning (TLS), however, they have not been fully analyzed on data captured by an MLS system. The paper presents our new approach, called the Optimum Single MLS Dataset method (OptD-single-MLS), which is an algorithm for MLS data reduction. The tests were carried out in two variants: (1) for raw sensory measurements and (2) for a georeferenced 3D point cloud. We found that the OptD-single-MLS method provides a good solution in both variants; therefore, the choice of the reduction variant depends only on the user.

MSNet: Multi-Scale Convolutional Network for Point Cloud Classification

Point cloud classification is quite challenging due to the influence of noise, occlusion, and the variety of types and sizes of objects. Currently, most methods mainly focus on subjectively designing and extracting features. However, the features rely on prior knowledge, and it is also difficult to accurately characterize the complex objects of point clouds. In this paper, we propose a concise multi-scale convolutional network (MSNet) for adaptive and robust point cloud classification. Both the local feature and global context are incorporated for this purpose. First, around each point, the spatial contexts of different sizes are partitioned as voxels of different scales. A voxel-based MSNet is then simultaneously applied at multiple scales to adaptively learn the discriminative local features. The class probability of a point cloud is predicted by fusing the features together across multiple scales. Finally, the predicted class probabilities of MSNet are optimized globally using the conditional random field (CRF) with a spatial consistency constraint. The proposed method was tested with data sets of mobile laser scanning (MLS), terrestrial laser scanning (TLS), and airborne laser scanning (ALS) point clouds. The experimental results show that the proposed method was able to achieve appreciable classification accuracies of 83.18%, 98.24%, and 97.02% on the MLS, TLS, and ALS data sets, respectively. The results also demonstrate that the proposed network has a strong generalization capability for classifying different kinds of point clouds under complex urban environments.

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

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

Automatic Detection and Classification of Pole-Like Objects for Urban Cartography Using Mobile Laser Scanning Data.

Mobile laser scanning (MLS) is a modern and powerful technology capable of obtaining massive point clouds of objects in a short period of time. Although this technology is nowadays being widely applied in urban cartography and 3D city modelling, it has some drawbacks that need to be avoided in order to strengthen it. One of the most important shortcomings of MLS data is concerned with the fact that it provides an unstructured dataset whose processing is very time-consuming. Consequently, there is a growing interest in developing algorithms for the automatic extraction of useful information from MLS point clouds. This work is focused on establishing a methodology and developing an algorithm to detect pole-like objects and classify them into several categories using MLS datasets. The developed procedure starts with the discretization of the point cloud by means of a voxelization, in order to simplify and reduce the processing time in the segmentation process. In turn, a heuristic segmentation algorithm was developed to detect pole-like objects in the MLS point cloud. Finally, two supervised classification algorithms, linear discriminant analysis and support vector machines, were used to distinguish between the different types of poles in the point cloud. The predictors are the principal component eigenvalues obtained from the Cartesian coordinates of the laser points, the range of the Z coordinate, and some shape-related indexes. The performance of the method was tested in an urban area with 123 poles of different categories. Very encouraging results were obtained, since the accuracy rate was over 90%.

A Density-Based Clustering Method for Urban Scene Mobile Laser Scanning Data Segmentation

The segmentation of urban scene mobile laser scanning (MLS) data into meaningful street objects is a great challenge due to the scene complexity of street environments, especially in the vicinity of street objects such as poles and trees. This paper proposes a three-stage method for the segmentation of urban MLS data at the object level. The original unorganized point cloud is first voxelized, and all information needed is stored in the voxels. These voxels are then classified as ground and non-ground voxels. In the second stage, the whole scene is segmented into clusters by applying a density-based clustering method based on two key parameters: local density and minimum distance. In the third stage, a merging step and a re-assignment processing step are applied to address the over-segmentation problem and noise points, respectively. We tested the effectiveness of the proposed methods on two urban MLS datasets. The overall accuracies of the segmentation results for the two test sites are 98.3% and 97%, thereby validating the effectiveness of the proposed method.

Extraction and Classification of Road Markings Using Mobile Laser Scanning Point Clouds

This study aims at building a robust method for semiautomated information extraction of pavement markings detected from mobile laser scanning (MLS) point clouds. The proposed workflow consists of three components: 1) preprocessing, 2) extraction, and 3) classification. In preprocessing, the three-dimensional (3-D) MLS point clouds are converted into radiometrically corrected and enhanced two-dimensional (2-D) intensity imagery of the road surface. Then, the pavement markings are automatically extracted with the intensity using a set of algorithms, including Otsu's thresholding, neighbor-counting filtering, and region growing. Finally, the extracted pavement markings are classified with the geometric parameters by using a manually defined decision tree. A study was conducted by using the MLS dataset acquired in Xiamen, Fujian, China. The results demonstrated that the proposed workflow and method can achieve 92% in completeness, 95% in correctness, and 94% in F -score.

DIGITAL TERRAIN MODELS FROM MOBILE LASER SCANNING DATA IN MORAVIAN KARST

Abstract. During the last ten years, mobile laser scanning (MLS) systems have become a very popular and efficient technology for capturing reality in 3D. A 3D laser scanner mounted on the top of a moving vehicle (e.g. car) allows the high precision capturing of the environment in a fast way. Mostly this technology is used in cities for capturing roads and buildings facades to create 3D city models. In our work, we used an MLS system in Moravian Karst, which is a protected nature reserve in the Eastern Part of the Czech Republic, with a steep rocky terrain covered by forests. For the 3D data collection, the Riegl VMX 450, mounted on a car, was used with integrated IMU/GNSS equipment, which provides low noise, rich and very dense 3D point clouds. The aim of this work is to create a digital terrain model (DTM) from several MLS data sets acquired in the neighbourhood of a road. The total length of two covered areas is 3.9 and 6.1 km respectively, with an average width of 100 m. For the DTM generation, a fully automatic, robust, hierarchic approach was applied. The derivation of the DTM is based on combinations of hierarchical interpolation and robust filtering for different resolution levels. For the generation of the final DTMs, different interpolation algorithms are applied to the classified terrain points. The used parameters were determined by explorative analysis. All MLS data sets were processed with one parameter set. As a result, a high precise DTM was derived with high spatial resolution of 0.25 x 0.25 m. The quality of the DTMs was checked by geodetic measurements and visual comparison with raw point clouds. The high quality of the derived DTM can be used for analysing terrain changes and morphological structures. Finally, the derived DTM was compared with the DTM of the Czech Republic (DMR 4G) with a resolution of 5 x 5 m, which was created from airborne laser scanning data. The vertical accuracy of the derived DTMs is around 0.10 m.

DIGITAL TERRAIN MODELS FROM MOBILE LASER SCANNING DATA IN MORAVIAN KARST

During the last ten years, mobile laser scanning (MLS) systems have become a very popular and efficient technology for capturing reality in 3D. A 3D laser scanner mounted on the top of a moving vehicle (e.g. car) allows the high precision capturing of the environment in a fast way. Mostly this technology is used in cities for capturing roads and buildings facades to create 3D city models. In our work, we used an MLS system in Moravian Karst, which is a protected nature reserve in the Eastern Part of the Czech Republic, with a steep rocky terrain covered by forests. For the 3D data collection, the Riegl VMX 450, mounted on a car, was used with integrated IMU/GNSS equipment, which provides low noise, rich and very dense 3D point clouds. <br><br> The aim of this work is to create a digital terrain model (DTM) from several MLS data sets acquired in the neighbourhood of a road. The total length of two covered areas is 3.9 and 6.1 km respectively, with an average width of 100 m. For the DTM generation, a fully automatic, robust, hierarchic approach was applied. The derivation of the DTM is based on combinations of hierarchical interpolation and robust filtering for different resolution levels. For the generation of the final DTMs, different interpolation algorithms are applied to the classified terrain points. The used parameters were determined by explorative analysis. All MLS data sets were processed with one parameter set. As a result, a high precise DTM was derived with high spatial resolution of 0.25 x 0.25 m. The quality of the DTMs was checked by geodetic measurements and visual comparison with raw point clouds. The high quality of the derived DTM can be used for analysing terrain changes and morphological structures. Finally, the derived DTM was compared with the DTM of the Czech Republic (DMR 4G) with a resolution of 5 x 5 m, which was created from airborne laser scanning data. The vertical accuracy of the derived DTMs is around 0.10 m.

Bag-of-visual-phrases and hierarchical deep models for traffic sign detection and recognition in mobile laser scanning data

This paper presents a novel algorithm for detection and recognition of traffic signs in mobile laser scanning (MLS) data for intelligent transportation-related applications. The traffic sign detection task is accomplished based on 3-D point clouds by using bag-of-visual-phrases representations; whereas the recognition task is achieved based on 2-D images by using a Gaussian-Bernoulli deep Boltzmann machine-based hierarchical classifier. To exploit high-order feature encodings of feature regions, a deep Boltzmann machine-based feature encoder is constructed. For detecting traffic signs in 3-D point clouds, the proposed algorithm achieves an average recall, precision, quality, and F-score of 0.956, 0.946, 0.907, and 0.951, respectively, on the four selected MLS datasets. For on-image traffic sign recognition, a recognition accuracy of 97.54% is achieved by using the proposed hierarchical classifier. Comparative studies with the existing traffic sign detection and recognition methods demonstrate that our algorithm obtains promising, reliable, and high performance in both detecting traffic signs in 3-D point clouds and recognizing traffic signs on 2-D images.

Predicting vegetation characteristics in a changing environment by means of laser scanning

Accurate and up-to-date information concerning vegetation characteristics is needed for decision-making from individual-tree-level management activities to the strategic planning of forest resources. Outdated information may lead to unbeneficial or even wrong decisions, at least when it comes to the timing of management activities. Airborne laser scanning (ALS) has so far been successfully used for applications involving detailed vegetation mapping because of its capability to simultaneously produce accurate information on vegetation and ground surfaces. The aim of this dissertation was to develop methods for characterizing vegetation and its changes in varying environments. A method called multisource single-tree inventory (MS-STI) was developed in substudy I to update urban tree attributes. In MSSTI stem map was produced with terrestrial laser scanning and by combining the stem map with predictors derived from ALS data it was possible to obtain improved estimates of diameter-at-breast height but also to produce new attributes such as height and crown size. Boat-based mobile laser scanning (MLS) data were employed in substudy II to map riverbank vegetation and identify changes. The overall classification accuracy of 73% was obtained, which is similar to accuracies found in other studies. With multi-temporal MLS data sets changes in vegetation were mapped year to year. In substudy III, open access ALS data were combined with multisource national forest inventory (NFI) data to investigate the drivers associated to wind damage. The special interest was in ALS-based predictors to map areas with wind disturbance and apply logistic regression to produce a continuous probability surface of wind predisposition to identify areas most likely to experience wind damage. The results demonstrated that a combination of ALS and multisource NFI in the modelling approach increased the prediction accuracy from 76% to 81%. The dissertation showed the capability of ALS and MLS for characterizing vegetation and mapping changes in varying environments. The developed applications could increase and expand the utilization of multi-temporal 3D data sets as well as increase data value. The results of this dissertation can be utilized in producing more accurate, diverse, and up-to-date information for decision-making related to natural resources.

Area-Based Approach for Mapping and Monitoring Riverine Vegetation Using Mobile Laser Scanning

Vegetation plays an important role in stabilizing the soil and decreasing fluvial erosion. In certain cases, vegetation increases the accumulation of fine sediments. Efficient and accurate methods are required for mapping and monitoring changes in the fluvial environment. Here, we develop an area-based approach for mapping and monitoring the vegetation structure along a river channel. First, a 2 × 2 m grid was placed over the study area. Metrics describing vegetation density and height were derived from mobile laser-scanning (MLS) data and used to predict the variables in the nearest-neighbor (NN) estimations. The training data were obtained from aerial images. The vegetation cover type was classified into the following four classes: bare ground, field layer, shrub layer, and canopy layer. Multi-temporal MLS data sets were applied to the change detection of riverine vegetation. This approach successfully classified vegetation cover with an overall classification accuracy of 72.6%; classification accuracies for bare ground, field layer, shrub layer, and canopy layer were 79.5%, 35.0%, 45.2% and 100.0%, respectively. Vegetation changes were detected primarily in outer river bends. These results proved that our approach was suitable for mapping riverine vegetation.

Derivation of tree skeletons and error assessment using LiDAR point cloud data of varying quality

The architecture of trees is of particular interest for 3D model creation in forestry and ecolocical applications. Terrestrial (TLS) and mobile laser scanning (MLS) systems are used to acquire detailed geometrical data of trees. Since 3D point clouds from laser scanning consist of large data amounts representing uninterpreted topographical information including noise and data gaps, an extraction of salient tree structures is important for further applications. We present a fully automated modular workflow for topological reliable reconstruction of tree architecture. Object-based point cloud processing such as branch extraction is combined with tree skeletonization. Branch extraction is performed using a segmentation procedure followed by segment-based analysis of form indices derived from eigenvector metrics. Extracted branch primitives are simplified and connected to line features during skeletonization. The modular workflow allows comprehensive parameter tests and error assessments that are used for a calibration of the module parameters with respect to various characteristics of the input data (e.g noise, scanning resolution, and the number of scan positions). The estimated parameter settings are validated using an exemplary MLS data set. The quality of input point cloud data, strongly influencing the quality of the skeleton results, can be improved by the presented branch extraction procedure. The potential for data improvement increases with increasing point densities. For our object-based appoach, we can show that the presence of erroneous structures and filtering artifacts have the strongest influence onto the quality of the derived skeletons. In contrast to traditional skeletonization approaches, the existance of data gaps has less influence onto the results.