Category: Research

Multi-look processing is a well-known technique actively used in the SAR research community. In the paper, a method is described for the extraction of the ground moving targets parameters from a sequence of SAR look images obtained with a single-antenna SAR. The method calls for the automatic road extraction from the multi-look SAR images. The proposed road detection and location algorithm is based on the Canny edge detector and the histograms of image gradient directions. The estimated road coordinates are used for unambiguous evaluation of the moving targets parameters, namely the velocity components and the coordinates

Autofocusing is one of the key steps in the high-resolution SAR imaging. In the paper several important improvements to the recently developed local-quadratic map-drift autofocus are proposed. Important SAR image preprocessing steps applied before the local Doppler rate errors estimation are described. The weighting scheme for the evaluation of residual cross-track acceleration components is developed. It is illustrated that accounting of the local image contrast leads to better estimation precision. Also it is shown that the usage of measured real antenna orientation angles results in more reliable estimation of the local Doppler rate errors. Performance of the trajectory reconstruction technique is tested with experimental data obtained with X-band airborne SAR system.

The accuracy of trajectory measurements is one of the crucial factors in high-resolution SAR imaging. Common navigation systems often do not fulfill the requirements that results in significant image quality degradation. In the paper, a new autofocus algorithm for the reconstruction of the SAR platform trajectory deviations is proposed. The approach is based on the estimation of the Doppler rate errors on a sequence of short-time intervals. The method is capable of estimation of time-varying and range-dependent phase error functions. The key steps of the developed technique are illustrated. The method is demonstrated on experimental data obtained with an X-band airborne SAR system.

Synthetic aperture radar is a popular instrument for high-resolution imaging. Usage of state-of-the-art image and signal processing solutions allows to significantly increase the efficiency of such systems. In the paper two important applications of computer vision algorithms are described. In particular, usage of local feature extraction algorithms for the radar image stitching. This results in the automatical panorama creation without usage of the navigation data. In addition, the road location approach based on the stroke width transform and contour analysis is proposed. Application of the developed methods is illustrated with real SAR data examples

Two modifications of the range-Doppler algorithm (RDA) have been proposed to solve problems of SAR platform motion instabilities. First, the multi-look processing based on the RDA with an extended Doppler bandwidth has been introduced for correction of radiometric errors. Second, the RDA has been modified to perform SAR image formation on short-time acquisition intervals to use it in a recently-developed local-quadratic map-drift autofocus (LQMDA) method. The performance of the methods is illustrated with experimental data obtained by airborne SAR systems

The quality of high-resolution SAR imaging is strictly related with the precision of the platform trajectory measurements. This is one of the crucial points in the case of image formation from light-weight aircrafts and UAV platforms, which undergone significant trajectory instabilities in real flight conditions. In the paper, a new approach for the residual range-dependent phase error restoration is proposed. According to the method, the cross-track aircraft accelerations are evaluated via a specifically developed weighted estimation procedure. The reconstructed residual trajectory deviations are used for the range-dependent motion errors compensation. The corresponding experimental results are also demonstrated.

In the paper, an X-band airborne SAR system developed and produced at the Institute of Radio Astronomy is described. The system is designed to be operated from light-weight aircrafts. Implemented hardware and real-time software solutions are discussed. Several original approaches to the post-processing of the recorded data are considered. Experimental results are also presented and discussed.

Presence of phase errors in radar data is a crucial problem for high-resolution SAR imaging. In the paper, a new idea for representation of an arbitrary phase error by its local approximations is proposed. Within such approach, a general analytical expression for the resolution degradation of multi-look SAR images is derived. The principle of local approximations is also used in a novel autofocusing approach that allows estimating and compensating of an arbitrary residual phase error function via measuring local quadratic phase errors.