This paper describes a method for automated detection of temporary cars in Mobile LiDAR point clouds. It consists of a segment-based classification of static cars and a comparison of data from two sensors to identify moving cars. Two... more
This paper describes a method for automated detection of temporary cars in Mobile LiDAR point clouds. It consists of a segment-based classification of static cars and a comparison of data from two sensors to identify moving cars. Two segmentation methods are used to extract the ground and group the above-ground points into objects. From each segmented object a number of features are extracted, and a classification strengthened by feature selection is performed to classify temporary cars. We evaluate the performance of two different classifiers trained with a training set including 117 temporary cars, and show classification accuracies of up to 92%. We also investigate a method for identifying moving cars based on the distance between corresponding segments in the point clouds captured by the two scanning sensors, and report an overall accuracy of 61%.
Research Interests:
In the past few years the number of applications that use 3D information of topographical objects increased rapidly. With the growing demand for 3D topographic data the need for automated 3D data acquisition also grows. Height information... more
In the past few years the number of applications that use 3D information of topographical objects increased rapidly. With the growing demand for 3D topographic data the need for automated 3D data acquisition also grows. Height information can be extracted from airborne or terrestrial acquisition methods, but can also be modelled as implicit semantic information. Adding height information to existing features is insufficient; additional features have to be acquired and existing features might get an extra dimension (surfaces can be converted to volumes, etc). The challenge is to produce semantical, geometrical and topological correct 3D topography. In this paper we describe the steps to acquire 3D topographic information. Special attention lies on the user requirements of 3D models. These requirements have been accomplished by information analysis at four major geo-information organizations in The Netherlands. The four cases describe the wishes and requirements for 3D data and modell...
Satellite radar interferometry (InSAR) techniques have been successfully applied for structural health monitoring of line-infrastructure such as railway. Limited by meter-level spatial resolution of Sentinel-1 satellite radar (SAR)... more
Satellite radar interferometry (InSAR) techniques have been successfully applied for structural health monitoring of line-infrastructure such as railway. Limited by meter-level spatial resolution of Sentinel-1 satellite radar (SAR) imagery and meter-level geolocation precision, it is still challenging to (1) categorize radar scatterers (e.g., persistent scatterers (PS)) and associate radar scatterers with actual objects along railways, and (2) identify unstable railway segments using InSAR Line of Sight (LOS) deformation time series from a single viewing geometry. In response to this, (1) we assess and improve the 3-D geolocation quality of Sentinel-1 derived PS using a 2-step method for PS 3-D geolocation improvement aided by laser scanning data; after geolocation improvement, we step-wisely classify railway infrastructure into rails, embankments and surroundings; (2) we recognize unstable rail segments by utilizing the (localized) differential settlement of rails in the normal dir...
Research Interests:
The labeling of point clouds is the fundamental task in airborne laser scanning (ALS) point clouds processing. Many supervised methods have been proposed for the point clouds classification work. Training samples play an important role in... more
The labeling of point clouds is the fundamental task in airborne laser scanning (ALS) point clouds processing. Many supervised methods have been proposed for the point clouds classification work. Training samples play an important role in the supervised classification. Most of the training samples are generated by manual labeling, which is time-consuming. To reduce the cost of manual annotating for ALS data, we propose a framework that automatically generates training samples using a two-dimensional (2D) topographic map and an unsupervised segmentation step. In this approach, input point clouds, at first, are separated into the ground part and the non-ground part by a DEM filter. Then, a point-in-polygon operation using polygon maps derived from a 2D topographic map is used to generate initial training samples. The unsupervised segmentation method is applied to reduce the noise and improve the accuracy of the point-in-polygon training samples. Finally, the super point graph is used ...
Research Interests:
3D Cadastre models capture both the complex interrelations between physical objects and their corresponding legal rights, restrictions, and responsibilities. Most of the ongoing research on 3D Cadastre worldwide is focused on... more
3D Cadastre models capture both the complex interrelations between physical objects and their corresponding legal rights, restrictions, and responsibilities. Most of the ongoing research on 3D Cadastre worldwide is focused on interrelations at the level of buildings and infrastructures. So far, the analysis of such interrelations in terms of indoor spaces, considering the time aspect, has not been explored yet. In The Netherlands, there are many examples of changes in the functionality of buildings over time. Tracking these changes is challenging, especially when the geometry of the spaces changes as well; for example, a change in functionality, from administrative to residential use of the space or a change in the geometry when merging two spaces in a building without modifying the functionality. To record the changes, a common practice is to use 2D plans for subdivisions and assign new rights, restrictions, and responsibilities to the changed spaces in a building. In the meantime,...
Research Interests:
Research Interests:
State-of-the-art indoor mobile laser scanners are now lightweight and portable enough to be carried by humans. They allow the user to map challenging environments such as multi-story buildings and staircases while continuously walking... more
State-of-the-art indoor mobile laser scanners are now lightweight and portable enough to be carried by humans. They allow the user to map challenging environments such as multi-story buildings and staircases while continuously walking through the building. The trajectory of the laser scanner is usually discarded in the analysis, although it gives insight about indoor spaces and the topological relations between them. In this research, the trajectory is used in conjunction with the point cloud to subdivide the indoor space into stories, staircases, doorways, and rooms. Analyzing the scanner trajectory as a standalone dataset is used to identify the staircases and to separate the stories. Also, the doors that are traversed by the operator during the scanning are identified by processing only the interesting spots of the point cloud with the help of the trajectory. Semantic information like different space labels is assigned to the trajectory based on the detected doors. Finally, the p...
Research Interests:
The data acquisition with Indoor Mobile Laser Scanners (IMLS) is quick, low-cost and accurate for indoor 3D modeling. Besides a point cloud, an IMLS also provides the trajectory of the mobile scanner. We analyze this trajectory jointly... more
The data acquisition with Indoor Mobile Laser Scanners (IMLS) is quick, low-cost and accurate for indoor 3D modeling. Besides a point cloud, an IMLS also provides the trajectory of the mobile scanner. We analyze this trajectory jointly with the point cloud to support the labeling of noisy, highly reflected and cluttered points in indoor scenes. An adjacency-graph-based method is presented for detecting and labeling of permanent structures, such as walls, floors, ceilings, and stairs. Through occlusion reasoning and the use of the trajectory as a set of scanner positions, gaps are discriminated from real openings in the data. Furthermore, a voxel-based method is applied for labeling of navigable space and separating them from obstacles. The results show that 80% of the doors and 85% of the rooms are correctly detected, and most of the walls and openings are reconstructed. The experimental outcomes indicate that the trajectory of MLS systems plays an essential role in the understandin...
Research Interests:
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... more
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 experime...
Research Interests:
Research Interests:
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... more
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 positi...
Research Interests:
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... more
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...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT Terrestrial laser scanning provides valuable information for building outlining, facade detection and building reconstruction. Especially mobile laser scanning (MLS) is considered as well suited to collect 3D point clouds from... more
ABSTRACT Terrestrial laser scanning provides valuable information for building outlining, facade detection and building reconstruction. Especially mobile laser scanning (MLS) is considered as well suited to collect 3D point clouds from building facades along road corridors for large areas. However, the completeness of facade representation in MLS has to be investigated in order to be able to draw conclusions about the usability of this kind of data sets for further applications such as building facade modelling. We investigate the detection rates of a fully automatic point cloud processing method for extracting building facades from MLS. The point cloud is segmented into planar regions, from which vertical structures are extracted. The detection rate is assessed by comparing the detected facade footprints with visible building outlines extracted from a digital cadastre map. The completeness of the extraction is investigated regarding the facade structure, length and the distance of the facades to the vehicle trajectory. It was found that the representation of facades extracted from MLS and the cadastral map might differ if very short facade parts (<2 m) representing jutties are present on the ground level floor. This leads to an underestimation of completeness. Moreover, it can be shown that there is a direct relationship between further characteristics, e. g. that long facades and facades near to the vehicle are more likely to be detected than others. Facades with a length between 10 m and 20m reach a completeness of 74%. Most facades are found in a distance of 10-20 m from the vehicle where the completeness ranges from 71% to 50%. The low completeness can be explained by occlusions from moving objects and vegetation, the facade structure, orientation and its complexity. The comparison with digital cadastral data shows that MLS is not that well suited for the detection of building facades as one might expect.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT: Mobile laser scanning acquires massive point clouds in urban areas to provide high resolution data for 3D city modelling. A workflow for detecting and modelling trees from point clouds is presented. Emphasis lies on data... more
ABSTRACT: Mobile laser scanning acquires massive point clouds in urban areas to provide high resolution data for 3D city modelling. A workflow for detecting and modelling trees from point clouds is presented. Emphasis lies on data reduction using an alpha shape approach. From the reduced point cloud the parameters are extracted to model the 3D trees using the Weber and Penn (1995) approach. The workflow is applied on two different sample data sets which were acquired with different mobile mapping systems and thus vary in quality ...
Research Interests:
Research Interests: Computer Science, Artificial Intelligence, Analytical Chemistry, Computer Vision, Sensor, and 15 moreRadar, Imaging, Housing, Medicine, Triangulation, Sensors, Transducers, Calibration, Laser Scanning, Error, Point Cloud, Equipment Design, Equipment Failure Analysis, Geographic Information Systems, and Electrical And Electronic Engineering
ABSTRACT Identifying damaged buildings after natural disasters such as earthquake is important for the planning of recovery actions. We present a segment-based approach to classifying damaged building roofs in aerial laser scanning data.... more
ABSTRACT Identifying damaged buildings after natural disasters such as earthquake is important for the planning of recovery actions. We present a segment-based approach to classifying damaged building roofs in aerial laser scanning data. A challenge in the supervised classification of point segments is the generation of training samples, which is difficult because of the complexity of interpreting point clouds. We evaluate the performance of three different classifiers trained with a small set of training samples and show that feature selection improves the training and the accuracy of the resulting classification. When trained with 50 training samples, a linear discriminant classifier using a subset of six features reaches a classification accuracy of 85%.
Research Interests:
Research Interests:
ABSTRACT Identifying damaged buildings after natural disasters such as earthquake is important for the planning of recovery actions. We present a segment-based approach to classifying damaged building roofs in aerial laser scanning data.... more
ABSTRACT Identifying damaged buildings after natural disasters such as earthquake is important for the planning of recovery actions. We present a segment-based approach to classifying damaged building roofs in aerial laser scanning data. A challenge in the supervised classification of point segments is the generation of training samples, which is difficult because of the complexity of interpreting point clouds. We evaluate the performance of three different classifiers trained with a small set of training samples and show that feature selection improves the training and the accuracy of the resulting classification. When trained with 50 training samples, a linear discriminant classifier using a subset of six features reaches a classification accuracy of 85%.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
ABSTRACT
Research Interests:
ABSTRACT This paper describes the generation and dissemination of a national three-dimensional (3D) dataset representing the virtual and landscape model. The 3D model is produced automatically by fusing a two-dimensional (2D) national... more
ABSTRACT This paper describes the generation and dissemination of a national three-dimensional (3D) dataset representing the virtual and landscape model. The 3D model is produced automatically by fusing a two-dimensional (2D) national object-oriented database describing the physical landscape and the national high-resolution height model of the Netherlands. Semantic constraints are introduced to correctly model 3D objects. Three areas from different regions in the Netherlands have been processed in order to develop, improve, and test the automatic generation of a national 3D city and landscape model. Specific attention has been paid to exceptional cases that may occur in a nationwide dataset. Based on the test results, the Kadaster, the national agency in the Netherlands responsible for the production of nation wide geo-information, decided that it is feasible to produce a national 3D city and landscape model that fulfills the specifications that were defined as part of this study. Future research is identified to make the results further ready for practice.
Research Interests:
ABSTRACT Airborne laser scanning data has proven to be a very suitable technique for the determination of digital surface models and is more and more being used for mapping and GIS data acquisition purposes, including the detection and... more
ABSTRACT Airborne laser scanning data has proven to be a very suitable technique for the determination of digital surface models and is more and more being used for mapping and GIS data acquisition purposes, including the detection and modeling of man-made objects or vegetation. The ...