Talks and Poster Presentations (with Proceedings-Entry):
"Applications of Image Based Measurement Systems and Geo-Referenced Data";
Talk: ESF TOPO-EUROPE Summer School,
Loen, Norwegen (invited);
- 2010-09-08; in: "ESF-TOPO-EUROPE Workshop and PhD Summer School on Detecting Landscape Change - Proceedings",
A. Beylich, K. Laute (ed.);
Rockfalls and landslides are major types of natural hazards worldwide that kill or injure a large number of individuals and cause very high costs every year. Risk assessment of such dangerous events requires an accurate evaluation of the geology, hydrogeology, morphology and interrelated factors such as environmental conditions and human activities. It is of particular importance for engineers and geologists to assess slope stability and dynamics in order to take appropriate, effective and timely measures against such events.
The main goal of this PhD-thesis is the development of a deformation interpretation tool for geo-risk assessment on the basis of a knowledge-based system. Currently this is done by human experts from geology and civil engineering, who are interpreting deformations on the basis of a large number of data records, documents and knowledge of different origin. The implementation of a knowledge-based system enables an automatic process of interpretation and determining of the risk potential. Towards this goal, we have carried out extensive knowledge acquisition and performed knowledge analysis with the help of several experts, leading to a rich knowledge base that builds on a number of influence factors, determined from various information sources (e.g., measurement data, expert knowledge, maps, etc.).
At the Vienna University of Technology (Institute of Geodesy and Geophysics), the interdisciplinary research project i-MeaS ("An Intelligent Image-Based Measurement System for Geo-Hazard Monitoring") has been launched with the purpose of research, develop and implement an interpretation tool for geo-risk objects. The system gives on-line information about ongoing deformations and supports issuing alerts in case of excessive deformation behavior.
Making conclusions about incidents is a not-trivial problem; by using artificial intelligence techniques, via the integration of a knowledge-based system, new directions are opened up. This new system is a complex intelligent system, working with several different data sets in real-time. Deformation measurement data will be delivered by a novel type of measurement system, which consists of two image-based sensors. Inside the captured images so-called interest points are detected. The calculation of the 3D coordinates is done by classical geodetic forward intersection. By means of such a high precision measurement system, 3D object points can be detected with an accuracy of about 2-3 mm (object distances up to 1000 m). Subsequently a geodetic deformation analysis can be performed that yields as a result deformation movement vectors, which constitute the input for later interpretation.
Based on the measured deformation vectors, a measurement preprocessing is performed (mainly clustering to detect areas of similar movement). On the basis of this information and additional data about velocity and orientation, some conclusions about the kind of occurring movement can be drawn. Additionally, data of different, heterogeneous sources, such as geodetic deformation measurements, geotechnical measurements, geological maps, geomorphological maps, in-situ investigations, and numerical modeling methods have to be included in such a system.
The concept of data interpretation is based on the "calculation" of risk factors for critical cause variables and on the elaboration of an interpretation for the deformation. Examples for cause variables can be precipitation, permafrost, critical slope angle, etc.
The challenging problem in developing such an alerting system is (1) to identify relevant factors and (2) subsequently to capture the interlinkage of these influence factors.
In our system, the process of risk assessment is divided into two steps: (1) the determination of the "Initial Risk Factor" and (2) the determination of the "Dynamic Risk Factor". The first step estimates the plausibility of an occurring moving event. Furthermore the zero state of interpretation and the observation is defined.
The second step is focused on the processing of the temporal development of the risk factor. Therefore additional data have to be included into the decision process, e.g., measured data captured by the image-based monitoring system. Measurement data represent the 3D object deformations (data is captured in defined time periods resulting in movement/displacement vectors). This system is also able to access local and global meteo data in real-time, which can be used by the dynamic system as a basis for deformation prediction.
Beside difficult technical requirements related to sensor and data fusion, the most challenging tasks in developing such a system is the implementation of the knowledge base and, in a preliminary step, the knowledge acquisition, especially the acquisition of knowledge from human experts. This problem was solved by using a two-step approach: in the first step, a single expert was consulted, while in the second step an extensive system evaluation by many experts was carried out and their feedback was incorporated into a system refinement. In order to estimate the initial risk factor, influence factors had to be identified whose values increase the likelihood of deformation. During a period of extensive discussions about the domain problem, more than thirty-five relevant influence factors were identified (e.g. vegetation, granual material, subsoil, pieces of rock, indicates, slope angle, slope profile, slip surface, material underground, saturation of soil, leaning trees, leaning rocks, crack, rock joint, joint orientation, insolation, permafrost, stone chips, frost-thaw-cycle, depth of movement, local temperature, etc.). About thirty of them are used for the determination of the "Initial Risk Factor".
For estimating the mentioned risk factor, we developed a knowledge-based system, adopting a rule-based approach, more specifically using production rules. This is because the connection between influence factors and possible causes or deformation behavior can be naturally formulated by rules, and this representation is more accessible to domain experts than other representations.
For the implementation, we have chosen JESS, which is a rule engine and scripting environment entirely written in JAVA.
The risk assessment is a difficult task where expert experience is required to obtain a reasonable solution. This is witnessed by the fact that there is no sole "correct" assessment for many of the different test cases, and expert opinions on the risk can vary for the same test case. The system can compete with the human experts; the differences between the result of the system on one side and the experts´ results vary in a similar way as the results vary between different experts. In order to further improve the quality of risk assessments of the system and to test its usability, we initiated an even broader evaluation where we asked additional experts for their opinion on our test cases. These new experts should bring in a fresh sight on the problem and the system because they have no information about the system and no training on it. The results obtained from first new probands are in accordance with the system and with the former evaluators. This also shows that the system interface is intuitive enough such that an untrained expert can easily use it without major difficulties.
An experimental prototype for risk assessment we developed shows good results for the completed "Initial Risk Factor" part, in where it behaves like a human expert.
Currently, the dynamic system (the second part of the final system) is under development - besides extending the existing rule set, further rule components (e.g., for updating risk factor, including meteorological data, etc.) will be added. Future work will include testing (collecting field data is targeted for the summer of 2010) and the integration of the knowledge base component into an on-line geo-risk management system.
rockfalls, landslides, surface detection, erosion rates
Electronic version of the publication:
Created from the Publication Database of the Vienna University of Technology.