Sander Oude Elberink

Accuracy and Resolution of Kinect Depth Data for Indoor Mapping Applications

Consumer-grade range cameras such as the Kinect sensor have the potential to be used in mapping applications where accuracy requirements are less strict. To realize this potential insight into the geometric quality of the data acquired by the sensor is essential. In this paper we discuss the calibration of the Kinect sensor, and provide an analysis of the accuracy and resolution of its depth data. Based on a mathematical model of depth measurement from disparity a theoretical error analysis is presented, which provides an insight into the factors influencing the accuracy of the data. Experimental results show that the random error of depth measurement increases with increasing distance to the sensor, and ranges from a few millimeters up to about 4 cm at the maximum range of the sensor. The quality of the data is also found to be influenced by the low resolution of the depth measurements.

Building Reconstruction by Target Based Graph Matching on Incomplete Laser Data: Analysis and Limitations

With the increasing point densities provided by airborne laser scanner (ALS) data the requirements on derived products also increase. One major application of ALS data is to provide input for 3D city models. Modeling of roof faces, (3D) road and terrain surfaces can partially be done in an automated manner, although many such approaches are still in a development stage. Problems in automatic building reconstruction lie in the dynamic area between assumptions and reality. Not every object in the data appears as the algorithm expects. Challenges are to detect areas that cannot be reconstructed automatically. This paper describes our contribution to the field of building reconstruction by proposing a target based graph matching approach that can handle both complete and incomplete laser data. Match results describe which target objects appear topologically in the data. Complete match results can be reconstructed in an automated manner. Quality parameters store information on how the model fits to the input data and which data has not been used. Areas where laser data only partly matches with target objects are detected automatically. Four datasets are analyzed in order to describe the quality of the automatically reconstructed roofs, and to point out the reasons why segments are left out from the automatic reconstruction. The reasons why these areas are left out include lack of data information and limitations of our initial target objects. Potential improvement to our approach is to include likelihood functions to the existence of topological relations.

Segment-Based Classification of Damaged Building Roofs in Aerial Laser Scanning Data

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%.

Automatic Extraction of Railroad Centerlines from Mobile Laser Scanning Data

In this paper, we describe the automatic extraction of centerlines of railroads. Mobile Laser Scanning systems are able to capture the 3D environment of the rail tracks with a high level of detail. Our approach first detects laser points that were reflected by the rail tracks, by making use of local properties such as parallelism and height in relation to neighboring objects. In the modeling stage, we present two approaches to determine the centerline location. The first approach generates center points in a data-driven manner by projecting rail track points to the parallel track, and taking the midpoint as initial center point. Next, a piecewise linear function is fitted through the center points to generate center points at a regular interval. The second approach models the rail track by fitting piecewise 3D track models to the rail track points. The model consists of a pair of two parallel rail tracks. The fitted pieces are smoothened by a Fourier series interpolation function. After that the centerline is implicitly determined by the geometric center of the pair of tracks. Reference data has been used to analyze the quality of our results, confirming that the position of the centerlines can be determined with an accuracy of 2–3 cm.

Change detection of trees in urban areas using multi-temporal airborne lidar point clouds

Light detection and ranging (lidar) provides a promising way of detecting changes of vegetation in three dimensions (3D) because the beam of laser may penetrate through the foliage of vegetation. This study aims at the detection of changes in trees in urban areas with a high level of automation using mutil-temporal airborne lidar point clouds. Three datasets covering a part of Rotterdam, the Netherlands, have been classified into several classes including trees. A connected components algorithm was applied first to group the points of trees together. The attributes of components were utilized to differentiate tree components from misclassified non-tree components. A point based local maxima algorithm was implemented to distinguish single tree from multiple tree components. After that, the parameters of trees were derived through two independent ways: a point based method using 3D alpha shapes and convex hulls; and a model based method which fits a Pollock tree model to the points. Then the changes were detected by comparing the parameters of corresponding tree components which were matched by a tree to tree matching algorithm using the overlapping of bounding boxes and point to point distances. The results were visualized and statistically analyzed. The difference of parameters and the difference of changes derived from point based and model based methods were both lower than 10%. The comparison of these two methods illustrates the consistency and stability of the parameters. The detected changes show the potential to monitor the growth and pruning of trees.

Detection and classification of changes in buildings from airborne laser scanning data

The difficulty associated with the Lidar data change detection method is lack of data, which is mainly caused by occlusion or pulse absorption by the surface material, e.g., water. To address this challenge, we present a new strategy for detecting buildings that are “changed”, “unchanged”, or “unknown”, and quantifying the changes. The designation “unknown” is applied to locations where, due to lack of data in at least one of the epochs, it is not possible to reliably detect changes in the structure. The process starts with classified data sets in which buildings are extracted. Next, a point-to-plane surface difference map is generated by merging and comparing the two data sets. Context rules are applied to the difference map to distinguish between “changed”, “unchanged”, and “unknown”. Rules are defined to solve problems caused by the lack of data. Further, points labelled as “changed” are re-classified into changes to roofs, walls, dormers, cars, constructions above the roof line, and undefined objects. Next, all the classified changes are organized as changed building objects, and the geometric indices are calculated from their 3D minimum bounding boxes. Performance analysis showed that 80%–90% of real changes are found, of which approximately 50% are considered relevant.

Pole-Like Road Furniture Detection and Decomposition in Mobile Laser Scanning Data Based on Spatial Relations

Road furniture plays an important role in road safety. To enhance road safety, policies that encourage the road furniture inventory are prevalent in many countries. Such an inventory can be remarkably facilitated by the automatic recognition of road furniture. Current studies typically detect and classify road furniture as one single above-ground component only, which is inadequate for road furniture with multiple functions such as a streetlight with a traffic sign attached. Due to the recent developments in mobile laser scanners, more accurate data is available that allows for the segmentation of road furniture at a detailed level. In this paper, we propose an automatic framework to decompose road furniture into different components based on their spatial relations in a three-step procedure: first, pole-like road furniture are initially detected by removing ground points and an initial classification. Then, the road furniture is decomposed into poles and attachments. The result of the decomposition is taken as a feedback to remove spurious pole-like road furniture as a third step. If there are no poles extracted in the decomposition stage, these incorrectly detected pole-like road furniture—such as the pillars of buildings—will be removed from the detection list. We further propose a method to evaluate the results of the decomposition. Compared with our previous work, the performance of decomposition has been much improved. In our test sites, the correctness of detection is higher than 90% and the completeness is approximately 95%, showing that our procedure is competitive to state of the art methods in the field of pole-like road furniture detection. Compared to our previous work, the optimized decomposition improves the correctness by 7.3% and 18.4% in the respective test areas. In conclusion, we demonstrate that our method decomposes pole-like road furniture into poles and attachments with respect to their spatial relations, which is crucial for road furniture interpretation.

Application of Template Matching for Improving Classification of Urban Railroad Point Clouds

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.

Space subdivision in indoor mobile laser scanning point clouds based on scanline analysis

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.

Context-Based Filtering of Noisy Labels for Automatic Basemap Updating From UAV Data

Unmanned aerial vehicles (UAVs) have the potential to obtain high-resolution aerial imagery at frequent intervals, making them a valuable tool for urban planners who require up-to-date basemaps. Supervised classification methods can be exploited to translate the UAV data into such basemaps. However, these methods require labeled training samples, the collection of which may be complex and time consuming. Existing spatial datasets can be exploited to provide the training labels, but these often contain errors due to differences in the date or resolution of the dataset from which these outdated labels were obtained. In this paper, we propose an approach for updating basemaps using global and local contextual cues to automatically remove unreliable samples from the training set, and thereby, improve the classification accuracy. Using UAV datasets over Kigali, Rwanda, and Dar es Salaam, Tanzania, we demonstrate how the amount of mislabeled training samples can be reduced by 44.1% and 35.5%, respectively, leading to a classification accuracy of 92.1% in Kigali and 91.3% in Dar es Salaam. To achieve the same accuracy in Dar es Salaam, between 50000 and 60000 manually labeled image segments would be needed. This demonstrates that the proposed approach of using outdated spatial data to provide labels and iteratively removing unreliable samples is a viable method for obtaining high classification accuracies while reducing the costly step of acquiring labeled training samples.