Building footprints provide essential information for mapping, disaster management, and other large-scale studies. Synthetic Aperture Radar (SAR) provides consistent data availability over optical images owing to its unique properties, which consequently makes it more challenging to interpret. Previous studies have demonstrated the success of automated methods using Convolutional Neural Networks to detect buildings in Very High Resolution (VHR) SAR images. However, the scarcity of such datasets that are available to the public can limit research progress in this field. We explored the impact of several data augmentation (DA) methods on the performance of building detection on a limited dataset of SAR images. Our results show that geometric transformations are more effective than pixel transformations. The former improves the detection of objects with different scale and rotation variations. The latter creates textural changes that help differentiate edges better, but amplifies non-object patterns, leading to increased false positive predictions. We experimented with applying DA at different stages and concluded that applying similar DA methods in training and inference showed the best performance compared with DA applied only during training. Some DA can alter key features of a building’s representation in radar images. Among them are vertical flips and quarter circle rotations, which yielded the worst performance. DA methods should be used in moderation to prevent unwanted transformations outside the possible object variations. Error analysis, either through statistical methods or manual inspection, is recommended to understand the bias presented in the dataset, which is useful in selecting suitable DAs. The findings from this study can provide potential guidelines for future research in selecting DA methods for segmentation tasks in radar imagery.

This article presents a new approach to the estimation of shift and rotation between two images from different kinds of imaging sensors. The first of the image is an orthophotomap that is created using optical sensors with georeference information. The second one is created utilizing a Synthetic Aperture Radar (SAR) sensor.The proposed solution can be mounted on a flying platform, and, during the flight, the obtained SAR images are compared with the reference optical images, and thus it is possible to calculate the shift and rotation between these two images and then the direct georeferencing error. Since both images have georeference information, it is possible to calculate the navigation correction in cases when the drift of the calculated trajectory is expected. The method can be used in platforms where there is no satellite navigation signal and the trajectory is calculated on the basis of an inertial navigation system, which is characterized by a significant error. The proposed method of estimating the navigation error utilizing Affine Scale-Invariant Feature Transform (ASIFT) and Structure from Motion (SfM) is described, and techniques for improving the quality of SAR imaging using despeckling filters are presented. The methodology was tested and verified using real-life SAR images. Differences between the results obtained for a few selected despeckling methods were compared and commented on. Deep investigation of the nature of the SAR imaging technique and noise creation character allows new algorithms to be developed, which can be implemented on flying platforms to support existing navigation systems in which trajectory error occurs.

this paper presents the concept of using the low-frequency array (LOFAR) radio telescope receiver for passive detection of air and space targets using illuminators of opportunity. The LOFAR radio telescope operates in a 110-250 MHz band, thus the focus is on VHF illuminators, such as DAB (digital radio) and DVB-T (digital television). For the detection of air targets, such as aircraft, illuminators of opportunity that are relatively close to the radio telescope are considered. Due to the Earth's surface curvature, the detection of space targets, such as orbital debris, requires the use of transmitters that are far away from the receiver. The paper presents a theoretical study of the detection range for aerial and space targets and preliminary results of the detection of air targets using nearby transmitters.

This paper presents selected results obtain under SARINA project. SARINA is a SAR-based Augmented Integrity Navigation Architecture proposed by authors of this paper. The main goal of the SARINA project was designing and assessing a novel aircraft Inertial Navigation System (INS) for missile and UAV that will make use of features extracted from SAR/InSAR imagery and on-board terrain landmark database in order to ensure robustness against uncompensated IMU errors due to possible GPS lack of.

In the paper experimental results of ISAR (Inverse Synthetic Aperture Radar) processing obtained with highresolution radar are presented. Targets under observation were ground moving vehicles, such as cars, trucks and tractors. The experiments were performed with a FMCW (Frequency- Modulated Continuous-Wave) radar operating at 94 GHz with almost 1 GHz of bandwidth. Due to the measurement scenario more typical for SAR (Synthetic Aperture Radar), than ISAR, i.e. targets moving along straight line crossing the antenna beam, algorithms usually applied for SAR processing have been used.

This paper presents an analysis of range migration and Doppler frequency rate-of-change for an airborne passive bistatic radar (PBR) utilizing the synthetic aperture radar (SAR) technique in the imaging of ground objects. The study was carried out for PBR using the Digital Video Broadcasting - Terrestrial (DVB-T) transmitter as an illuminator of opportunity. The results obtained have been verified using both simulated and real data gathered by a PBR mounted on an airborne platform.