Year: 2010
Type: conference-paper
Abstract: The development of reliable change detection techniques from remote sensing data is one of the main challenges in urban growth and change monitoring research.One of the main open issues consists in automatizing the detection of changes whose interpretation has remained up-to-now visual in most operational applications in remote sensing.When dealing with urban areas, one possibility to cope with the automatic growth monitoring is the exploitation of the height information relative to the different man-made objects that exist in the scene.In fact, the comparison of Digital Surface Models (DSMs), acquired at different epochs, should provide a valuable information about the 3D urban changes occurred in the studied area.Nevertheless, while most of the changes are optically visible and easily detectable by an expert user, automatic processes are difficult to achieve due to the problems of co-registration and the significant height difference that may occur between DSMs acquired from different sensors and/or generated by different algorithms.This results in the detection of virtual or irrelevant changes.This article proposes two semi-automatic methods for 3D change detection using DSMs obtained from different sources.While the first method is based on the simple subtraction of DSMs from two epochs, the second one consists in comparing the classification maps of these two DSMs, through class-for-class differencing.In both cases, adaptative post-processing steps have been introduced in order to distinguish real from virtual changes.Evaluations of the proposed approaches have been carried out to detect the 3D changes that have occurred in the city center of Munich in Germany from 2003 to 2005.
Source: elib (German Aerospace Center)
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W29544197
