Auflistung nach Autor:in "Annanias, Yves"
1 - 2 von 2
Treffer pro Seite
Sortieroptionen
- KonferenzbeitragClassification of Large and Heterogeneous LiDAR Data Sets(INFORMATIK 2023 - Designing Futures: Zukünfte gestalten, 2023) Kaubukowski, Kenn; Nsonga, Baldwin; Kretzschmar, Vanessa; Annanias, Yves; Wiegreffe, Daniel; Hlawitschka, MarioIn the context of the renaturation of discontinued open-cast mines, the interactive visualization analysis of three-dimensional LiDAR provides a comprehensive overview for planning the subsequent use of these areas. When analyzing the measured point clouds, it is beneficial to classify the points, enabling the user to filter objects such as vehicles and buildings. Most current classification techniques for LiDAR data rely on good ground truth and work on specific types of measurements and resolutions. In this work, we present a semantic classification workflow for LiDAR data with highly heterogeneous acquisition and storage parameters. We apply the workflow to a freely available LiDAR data set and showcase the resulting classification. Our method is shown to reliably classify not only large-scale objects, such as buildings, but also small-scale objects, such as power lines and street lights.
- TextdokumentReport on the Correction of Erroneous Geometry Data in Land Reuse Projects(INFORMATIK 2021, 2021) Annanias, Yves; Wahsner, Marc; Scheuermann, Gerik; Wiegreffe, DanielLand Reuse processes are large planning and decision-making processes based on a large amount of geographic data. Therefore, it is essential that this data is as accurate as possible. However, errors can occur during the creation of the data and not all of them are directly noticeable. We report here what errors we have encountered while working with this geographic data, what problems they can cause, and how we have fixed them. Since the correction can be very time-consuming with the enormous amount of data, we have focused on an automatic correction. Not all of this data can be corrected this way, for the rest, we briefly indicate a procedure to support and simplify the manual correction.