Peter Dorninger

A Comprehensive Automated 3D Approach for Building Extraction, Reconstruction, and Regularization from Airborne Laser Scanning Point Clouds

Three dimensional city models are necessary for supporting numerous management applications. For the determination of city models for visualization purposes, several standardized workflows do exist. They are either based on photogrammetry or on LiDAR or on a combination of both data acquisition techniques. However, the automated determination of reliable and highly accurate city models is still a challenging task, requiring a workflow comprising several processing steps. The most relevant are building detection, building outline generation, building modeling, and finally, building quality analysis. Commercial software tools for building modeling require, generally, a high degree of human interaction and most automated approaches described in literature stress the steps of such a workflow individually. In this article, we propose a comprehensive approach for automated determination of 3D city models from airborne acquired point cloud data. It is based on the assumption that individual buildings can be modeled properly by a composition of a set of planar faces. Hence, it is based on a reliable 3D segmentation algorithm, detecting planar faces in a point cloud. This segmentation is of crucial importance for the outline detection and for the modeling approach. We describe the theoretical background, the segmentation algorithm, the outline detection, and the modeling approach, and we present and discuss several actual projects.

On-the-job detection and correction of systematic cyclic distance measurement errors of terrestrial laser scanners

Abstract Application of terrestrial laser scanning for engineering and documentation projects requiring high accuracy is often prohibitive due to the error budget of terrestrial laser scanners (TLS) which is worse than that of alternative instruments such as total stations. However, by separating the errors of TLS into random and systematic components, the influence of both can be reduced significantly. Approaches for reducing the influence of random noise rely on averaging. One method is introduced briefly as it is relevant for the task of the proposed systematic error removal. We discuss the correction of cyclic distance measurement errors of TLS in detail. For the investigated scanner, these cyclic errors are the most dominant systematic errors. The proposed approach is based on an on-the-job determination of a distance correction function which allows reducing the magnitude of the occurring distance errors by a factor of ten, reducing the systematic errors to less than 0.5 mm. The massive amount of data acquired by TLS is exploited using robust statistical methods. The algorithm developed relies on small, flat surfaces found automatically in the scene. Hence, the method can be applied on engineering project data without requiring additional laboratory experiments. Therefore, it is not essential that the errors are constant over time. By the introduction of on-the-job correction procedures as the one proposed, the systematic errors of TLS can be decreased significantly. This makes them more suitable to be used as an alternative to single point measurement devices.

Classification of airborne laser scanning point clouds based on binomial logistic regression analysis

This article presents a newly developed procedure for the classification of airborne laser scanning (ALS) point clouds, based on binomial logistic regression analysis. By using a feature space containing a large number of adaptable geometrical parameters, this new procedure can be applied to point clouds covering different types of topography and variable point densities. Besides, the procedure can be adapted to different user requirements. A binomial logistic model is estimated for all a priori defined classes, using a training set of manually classified points. For each point, a value is calculated defining the probability that this point belongs to a certain class. The class with the highest probability will be used for the final point classification. Besides, the use of statistical methods enables a thorough model evaluation by the implementation of well-founded inference criteria. If necessary, the interpretation of these inference analyses also enables the possible definition of more sub-classes. The use of a large number of geometrical parameters is an important advantage of this procedure in comparison with current classification algorithms. It allows more user modifications for the large variety of types of ALS point clouds, while still achieving comparable classification results. It is indeed possible to evaluate parameters as degrees of freedom and remove or add parameters as a function of the type of study area. The performance of this procedure is successfully demonstrated by classifying two different ALS point sets from an urban and a rural area. Moreover, the potential of the proposed classification procedure is explored for terrestrial data.

High-resolution 3D surface modeling of a fossil oyster reef

The world’s largest fossil oyster reef, formed by the giant oyster Crassostrea gryphoides and located in Stetten (north of Vienna, Austria), is studied in this article. Digital documentation of the unique geological site is provided by terrestrial laser scanning (TLS) at the millimeter scale. Obtaining meaningful results is not merely a matter of data acquisition with a suitable device; it requires proper planning, data management, and postprocessing. Terrestrial laser scanning technology has a high potential for providing precise 3D mapping that serves as the basis for automatic object detection in different scenarios; however, it faces challenges in the presence of large amounts of data and the irregular geometry of an oyster reef. We provide a detailed description of the techniques and strategy used for data collection and processing. The use of laser scanning provided the ability to measure surface points of 46,840 (estimated) shells. They are up to 60-cm-long oyster specimens, and their surfaces are modeled with a high accuracy of 1 mm. In addition, we propose an automatic analysis method for identifying and enumerating convex parts of shells. Object surfaces were detected with a completeness of 69% and a correctness of over 75% by means of a fully automated workflow. Accuracy of 98% was achieved in detecting the number of objects. In addition to laser scanning measurements, more than 300 photographs were captured, and an orthophoto mosaic was generated with a ground sampling distance (GSD) of 0.5 mm. This high-resolution 3D information and the photographic texture serve as the basis for ongoing and future geological and paleontological analyses. Moreover, they provide unprecedented documentation for conservation issues at a unique natural heritage site.