Assessment System of Gis-objects Using Multi-temporal Ima Gery for Near-realtime Disaster Management
Year: 2010
Type: article
Abstract: In this paper, a damage assessment system of GIS-objects such as roads and buildings after natural disasters is presente d. The main contribution is the integration and exploitation of multi- temporal imagery leading to a more robust assessment of infrastructural objects. In addition, the chronological development of the assessed objects is investigated. The multivariate alteration dete ction method is used to detect changes between different time points in conjunction with the classification of different changes realized vi a Gaussian mixture models. Further accessorily introduced information are derived from GIS, in particular DEM belief functions. The strategy of the proposed approach is the combination of the computed probabilities using individual appropriate methods. The goal of t he system is the assignment of GIS-objects into different damage assessment categories as intact or not intact/destroyed using the fused information from multi-temporal multi-sensorial data. The system is tested at a test scenario assessing roads concerning their tra fficability. The results show the improvement of the damage assessment system after the integration of multi-temporal information. 1. INTRODUCTION In this paper, an assessment system of GIS-objects is presented using multi-sensorial and multi-temporal imagery after natural disasters. The focus of this article is the multi-temporal c omponent, because the integration of imagery from different time points into an assessment system has several advantages: Firstly, multi-temporal images provide the opportunity to monitor natural disaster chronologically during a period of time, not on ly at a specific time point. Secondly, the assessment of the GIS-obj ects at the time point t2 can be improved using the results from time point t1. Another focus of this article is the automatic information r etrieval from imagery being relevant for rescue teams after natural disasters. Information on the status of the infrastructure after disasters is essential to guarantee an effective and fast disaster man agement. Therefore, the emphasis of this article is the development of automated methods assessing infrastructural objects such as roads concerning their functionality.
Authors: Daniel Frey, Matthias Butenuth
Institution: Technical University of Munich
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2181557164