Information extraction and visualization: Development of novel approaches and tools for the automatic extraction of relevant change information from 4D PLS data and for the visualization of extracted change information for end users.
AImon5.0 - Real-time monitoring of gravitational mass movements for critical infrastructure risk management with AI-assisted 3D metrology
(Echtzeitüberwachung gravitativer Massenbewegungen zum Risikomanagement kritischer Infrastrukturen mit KI-unterstützter 3D-Metrologie)
This website can be accessed via a short URL: www.uni-heidelberg.de/aimon.
The Earth's surface is constantly changing. Climate change is altering environmental conditions - for example, more intense and prolonged precipitation is causing more frequent landslides or rockfalls. Such events do not only affect the local population but also central and critical infrastructure. A key tool in integrated risk management is the availability of 4D geo-information. The information is collected through continuous monitoring in near real time.
The current state of the art is fixed and autonomous permanent laser scanner (PLS) systems. PLS systems provide huge amounts of data (billions of measurements per day). In order to make PLS systems and 4D (3D + time) analysis methods available for operational use and to limit the big 4D data to the relevant information for decision makers, a new interface between the information needs of the application and the 4D acquisition and analysis is required.
For the first time, this will make it possible to use state-of-the-art PLS systems in operational risk monitoring and, with the help of artificial intelligence methods, to find and evaluate specific relevant events (e.g. activation of a slope area) in the huge amounts of data, to follow them in continuous monitoring and to automatically identify new events.
In AImon5.0, our open-source frameworks HELIOS++ and py4dgeo will be combined to enhance current approaches for the automatic extraction and visualization of change from 4D PLS data.
AImon5.0 is an interdisciplinary collaboration project of the 3DGeo research group with DMT GmbH & Co. KG (project leader), TU Munich, Deutsche Bahn Netz AG and Landesamt für Geologie und Bergbau Rheinland Pfalz. The expertise of all partners will be combined to close the gap between research and application of operational risk monitoring.
In this project of the 3DGeo research group Heidelberg the following central research objectives will be addressed:
- Identify which computer-based methods for automatic information extraction from 4D PLS data and their visualisation are particularly suitable for operational use and how these may have to be adapted and extended in order to deliver reliable and timely results.
- Investigate and combine two complementary approaches (rule-based 4D change analysis and data-driven approaches using AI and virtual laser scanning) for the automatic extraction of change information.
- Find out which levels of abstraction and forms of visualization are best suited for certain tasks and also given reaction times.
AImon5.0 will thereby close the gap between existing 4D analysis methods and an operational implementation in a real-time assistance system.
Project partners and their research focuses in the project
DMT GmbH & Co. KG serves as the coordinator of the interdisciplinary research project. Components and services developed by the individual partners will be integrated into a single system.
System implementation: Implementation and demonstration of 4D PLS system and development of a web-based information tool for monitoring, data management and visualization of results in near-realtime.
Data quality and analysis: Development of methods for the quantification and reduction of uncertainty in 4D PLS data and for the detection and geometric parametrization of change.
- HELIOS++: Heidelberg LiDAR Operations Simulator.
- py4dgeo: Open source Python library for geographic change analysis in 4D point cloud data.
- 4D Objects-By-Change: Change Analysis of Natural Surfaces using 3D Time Series.
- CharAct4D: Unravelling Landscape Dynamics via Automatic Characterization of Surface Activity using Geographic 4D Monitoring.
- VirtuaLearn3D: Virtual Laser Scanning for Machine Learning Algorithms in Geographic 3D Point Cloud. Analysis
- Winiwarter, L., Anders, K., Czerwonka-Schröder, D. & Höfle, B. (2023): Full four-dimensional change analysis of topographic point cloud time series using Kalman filtering. Earth Surface Dynamics. Vol. 11 (4), pp. 593-613. DOI: 10.5194/esurf-11-593-2023.
- Anders, K., Winiwarter, L., Höfle, B. (2022): Automatic Extraction and Characterization of Natural Surface Changes from Near-Continuous 3D Time Series using 4D Objects-By-Change and Kalman Filtering. EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022. DOI: 10.5194/egusphere-egu22-4225.
- Anders, K., Winiwarter, L., Schröder, D. & Höfle, B. (2022): Integration of Kalman Filtering of Near-Continuous Surface Change Time Series into the Extraction of 4D Objects-By-Change. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLIII-B2-2022, 973--980. DOI: 10.5194/isprs-archives-XLIII-B2-2022-973-2022.
- Anders, K., Winiwarter, L., Höfle, B. (2022): Improving change analysis from near-continuous 3D time series by considering full temporal information. IEEE Geoscience and Remote Sensing Letters, 19. DOI: 10.1109/LGRS.2022.3148920.
- Hulskemper, D., Anders, K., Antolínez, J.A.Á., Kuschnerus, M., Höfle, B., Lindenbergh, R. (2022): Characterization of Morphological Surface Activities derived from Near-Continuous Terrestrial LiDAR Time Series. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLVIII-2/W2-2022, 53-60. DOI: 10.5194/isprs-archives-XLVIII-2-W2-2022-53-2022.
- Vos, S., Anders, K., Kuschnerus, M., Lindenbergh, R., Höfle, B., Aarninkhof, S., de Vries, S. (2022): A high-resolution 4D terrestrial laser scan dataset of the Kijkduin beach-dune system, The Netherlands. Scientific Data, 9 (191). DOI: 10.1038/s41597-022-01291-9.
- Anders, K., Winiwarter, L., Höfle, B. (2021): Improving change analysis from near-continuous 3D time series by considering full temporal information [Data and Source Code]. heiDATA, V1. DOI: 10.11588/data/1L11SQ.
- Anders, K., Winiwarter, L., Lindenbergh, R., Williams, J.G., Vos, S.E. & Höfle, B. (2020): 4D objects-by-change: Spatiotemporal segmentation of geomorphic surface change from LiDAR time series. ISPRS Journal of Photogrammetry and Remote Sensing. Vol. 159, pp. 352-363. DOI: 10.1016/j.isprsjprs.2019.11.025.
- Anders, K., Winiwarter, L., Mara, H., Lindenbergh, R. C., Vos, S. E. & Höfle, B. (2021): Influence of Spatial and Temporal Resolution on Time Series-Based Coastal Surface Change Analysis using Hourly Terrestrial Laser Scans. ISPRS Annals of Photogrammy, Remote Sensing and Spatial Information Science, V-2-2021, 137–144. DOI: 10.5194/isprs-annals-V-2-2021-137-2021.
- Anders, K., Winiwarter, L., Mara, H., Lindenbergh, R., Vos, S.E. & Höfle, B. (2021): Fully automatic spatiotemporal segmentation of 3D LiDAR time series for the extraction of natural surface changes. ISPRS Journal of Photogrammetry and Remote Sensing. Vol. 173, pp. 297-308. DOI: 10.1016/j.isprsjprs.2021.01.015. 10.1016/j.isprsjprs.2021.01.015.
- Höfle, B., Canli, E., Schmitz, E., Crommelinck, S., Hoffmeister, D. & Glade, T. (2016): 4D Near Real-Time Environmental Monitoring Using Highly Temporal LiDAR. In: Geophysical Research Abstracts. Vol. 18, pp. 1-1. DOI: 10.1016/j.isprsjprs.2021.01.015.
- Canli, E., Höfle, B., Hämmerle, M., Thiebes, B. & Glade, T. (2015): Permanent 3D laser scanning system for an active landslide in Gresten (Austria). In: Geophysical Research Abstracts. Vol. 17, pp. 1-1.
- Canli, E., Thiebes, B., Höfle, B. & Glade, T. (2015): Permanent 3D Laser Scanning System for Alpine Hillslope Instabilities. In: 6th International Conference on Debris-Flow Hazards Mitigation (DFHM): Mechanics, Prediction and Assessment. Tsukuba, Japan pp. 1-1.