SAR sensors play an important role in different fields of remote sensing. One of these is Automatic Target Recognition (ATR). In this paper, a new dataset for ATR is introduced, consisting of five classes of passenger cars imaged by SmartRadar of Hensoldt Sensors GmbH. The basic characteristics of the dataset and some details of the measurement campaign are provided. The second part of the paper deals with the creation of a sufficiently large database of training samples to train a Convolutional Neural Network (CNN) to classify these cars. Since training data are not as readily available as in the EO case, training data are simulated using the CohRaS SAR simulator of Fraunhofer IOSB, which is also briefly described. The basic setup of the CNN used for the classification task is outlined and some issues arising in the classification of the training data are discussed. The paper also contains some very preliminary classification results using the CNN and the simulated training data, and a discussion of these results.
Damage detection after natural disasters is one of the remote sensing tasks in which Synthetic Aperture Radar (SAR) sensors play an important role. Since SAR is an active sensor, it can record images at all times of day and in all weather conditions, making it ideally suited for this task. While with the newer generation of SAR satellites such as TerraSAR-X or COSMOSkyMed amplitude change detection has become possible even for urban areas, interferometric phase change detection has not been published widely. This is mainly because of the long revisit times of common SAR sensors leading to temporal decorrelation. This situation has changed dramatically with the advent of the TanDEM-X constellation, which can create single-pass interferograms from space at very high resolutions, avoiding temporal decorrelation almost completely. In this paper the basic structures that are present for any building in InSAR phases, i.e. layover, shadow, and roof areas, are examined. Approaches for their extraction from TanDEM-X interferograms are developed using simulated SAR interferograms. The extracted features of the building signature will in the future be used for urban change detection in real TanDEM-X High Resolution Spotlight interferograms.
The current generation of SAR satellites such as TerraSAR-X, TanDEM-X and COSMO-SkyMed provide resolutions below one meter, permitting the detailed analysis of urban areas while covering large zones. Furthermore, as they are deployable independently of daylight and weather, such remote sensing SAR data are particularly popular for purposes such as rapid damage assessment at building level after a natural disaster.
The purpose of our study is the investigation of techniques for the detection of changes based on one pre-event and one post-event SAR amplitude image. We provide a comparison of several methods for detecting changes in urban areas. Especially, changes at building locations are looked for. We analyzed two areas affected differently in detail. First, a suburban area of Paris, France, was considered due to changes caused by an urbanization project. Here, we have two TanDEM-X acquisitions available, before (November 4, 2012) and after (May 10, 2013) the changes.
Second, we investigated changes that happened in Kathmandu, Nepal, after the April 25, 2015 earthquake. For this analysis, we have two TerraSAR-X acquisitions, one before (October 13, 2013) and one immediately after (April 27, 2015) the earthquake. Both areas differ by the building types, the image resolution and the available reference, which makes it an interesting challenge.
In this paper, we compare six different methods for change detection. The investigated methods contain both standard criteria such as Log Ratio, Kullback-Leibler and the Difference of Entropies detector, and methods developed by the authors such as a Log Ratio combined with an Alternating Sequential Filter. All change detection results are presented and discussed by considering the available ground truth.
This second part of the paper about the creation and compilation of the Christchurch, New Zealand, dataset for the detection
of earthquake damages in urban areas deals with the extraction of additional information from the 3D model that can
aid in the detection of destructions. This includes the creation of a height image of the scene, shadow and layover masks
and using a modified version of the SAR simulator CohRaS® to simulate masks of the expected location of specular reflections
in the real TerraSAR-X scene after the earthquake. The algorithms used for the extraction of these data sets and
some ideas for their application for the damage detection task are discussed and first preliminary results are shown.
As the introducing first part of this paper, the data set of Christchurch, New Zealand, is outlined with regard to its purpose: the detection of earthquake damages. The aim is to produce simulated SAR images that are realistic enough to function successfully as pre-event images in a change detection effort. To this end, some modifications to the input 3D city model are introduced and discussed. This includes the use of a GIS map, for a realistic modelling of the radiometric variety, and the insertion of high vegetation to the model, so as to achieve a realistic occlusion of building corners. A detailed description of the impact, these modifications have on the simulation, is given and a comparison between the simulations and corresponding real data is drawn.
At Fraunhofer IOSB the SAR simulator suite CohRaSS (Coherent Raytracing SAR Simulator) dedicated to different,
sometimes contradictory purposes is being developed. These include the simulation of very large scenes at high resolution
for scene analysis purposes, the simulation of large quantities of training chips for classification and the very fast but
less realistic simulation of scenes for use in the training of image analysts. These tasks have very different requirements
for the simulation that cannot be met by one single program. Thus different, custom-tailored approaches for each of these
tasks are being developed. This paper deals with the main aspects concerning the simulation of training chips for ATR
and the simulation of large scenes at very high resolution. Special focus is set on the different approaches used for these
tasks from a computational point of view. For both simulators, sample simulated images are shown.
Simulated SAR images can be used for a wide variety of purposes such as classification, object recognition or the education
of SAR image analysts. In this paper, a SAR simulation tool based on an extended raytracing approach is introduced.
The approach allows for a coherent simulation of the most important SAR image features. Another aspect is the
modeling of material properties necessary for the creation of synthetic SAR images that look realistic. Also, the image
formation process is described in detail since it has immense impact on the overall appearance of the final image. The
different steps of the simulation process and their influence on the appearance of the final simulated image are demonstrated
with simulated images of a simple building. Then, examples for the simulation of large scale scenes at high resolution
SAR is a remote sensing technique capable to deliver actual data at any time and under bad weather conditions. Before
launch of TerraSAR-X, RADARSAT-2, or COSMO-SkyMed, the rather coarse resolution of operational SAR satellite
systems allowed an analysis of spaceborne SAR data in case of disaster management only for medium scale products.
The new generation of spaceborne SAR satellites permits a more detailed analysis at the object level even for urban
areas, which was before restricted to airborne SAR sensors. Change detection in SAR images is an important field of
research. In general, the appearance of objects in SAR images strongly depends on the viewing angle and look direction.
This makes a comparison of images on a pixel level difficult. The changeover from pixel- to object level leads to the
possibility, to look for object-features that are more stable concerning different imaging constellations. Bridges are keyelements
of man made infrastructure. In this paper the appearance of bridges in SAR data is analyzed and features are
derived that are exploitable for change detection. Here the focus is on analysis at the object level to derive features that
are either stable concerning the imaging constellations or that can be predicted based on a given imaging constellation.
Thereby, the usage of different sensors will be possible to achieve the goal of real time information. The investigations
are supported by simulations, which allow the creation of SAR images for different imaging constellations, bridge
materials, and even for situations with destroyed bridges.
Operational SAR satellite systems such as ENVISAT-ASAR and RADARSAT-1 deliver image data of a rather coarse
resolution, which allows the recognition or feature extraction only for large man-made objects. State of the art airborne
SAR sensors on the other hand provide spatial resolution in the order well below a half meter. In such data many features
of urban objects can be identified and used for recognition. Core elements of man-made infrastructure are bridges. In
case of bridges over water, the oblique side looking imaging geometry of SAR sensors may lead to special signature in a
SAR image depending on the aspect. In this paper, the appearance of bridges over water in SAR data is discussed.
Geometric constraints concerning the changing of this signature are investigated using simulation techniques based on an
adapted ray tracing. Furthermore, an approach is presented to detect bridges over water and to derive object features
from spaceborne and airborne SAR images in the context of disaster management. RADARSAT-1 data with a spatial
resolution of about 9 m as well as high-resolution airborne SAR data of geometric sampling distance better than 40 cm