Change detection in time series of high resolution SAR satellite images

Abstract

In the last few years, change detection based on remote sensing data has become a highly frequented field of research with multiple applications for practical use. To detect changes between temporarily different satellite images is of interest for example in terms of urban monitoring and disaster management. The approach presented in this paper allows the fully automatic detection of small-scaled changes (e.g. vehicles or construction sites) in time series of SAR amplitude image data. To create a robust method, only one single parameter encoding the size of the detected changes has to be set by the operator. Furthermore, first steps concerning the categorization of the detected changes are presented. As dataset, a time series of high resolution SAR images acquired by the German satellite TerraSAR-X was used. The time span of this time series, acquired in ascending and descending orbit, is about half a year

Using Morphological Differential Attribute Profiles for Change Categorization in High Resolution Sar Images

“…It is also shown that the combination of the two attributes leads to a more precise result. The detected change segments given in Figure 3 (upper half, right column) represent changes of comparable category (Boldt et al, 2012).…”

“…The change detection method described in (Boldt et al, 2012) requires an amplitude image pair as data input. From this image pair, a maximum ratio image is calculated.…”

“…(Bovolo et al, 2012), (Acito et al, 2012)) without analyzing the category of the detected changes. Providing support for this lack, different features for change categorization were considered in (Boldt et al, 2012). As one of these features, the morphological differential attribute profiles (DAPs) were presented.…”

“…Since pixel-based methods are not well-suited for investigating HR imagery, object-based methods are preferred. In (Boldt et al, 2012), a fully automatic object-based method for change detection in HR SAR image pairs was presented. This method relies on a morphological alternating sequential filtering step that leads to binary change segments after thresholding.…”

“…Furthermore, it is of special interest to derive information on the category of detected changes is of special interest. In (Boldt et al, 2012), a fullyautomatic change detection method based on a morphological filtered ratio image was presented. This filter step is accomplished by an alternating sequential filter (ASF), which supports the knowledge-driven analysis of the scene.…”