SAR Imaging: An Autofocusing Method for Improving Image Quality and MFS Image Classification Technique

Abstract

In the first part of this paper several aspects of the SAR imaging are presented. Firstly, the mathematical theory and methodology for generating SAR synthetic backscattered data are developed. The simulated target is a ship, which is located on the sea surface. A two-dimensional and a three-dimensional target (ship) implementations are included in the simulations. Both cases of airborne and spaceborne SAR are simulated. Furthermore, the case of varying target scattering intensity is presented. In addition an application of an autofocusing algorithm, previously developed by the authors for the case of Inverse Synthetic Aperture Radar (ISAR) and Synthetic Aperture Radar (SAR) geometry for simulated data, is presented here for the case of real-field radar data, provided to us by SET 163 Working Group. This algorithm is named “CPI-split-algorithm”, where CPI stands for “Coherent Processing Interval”. Numerical results presented in this paper show the effectiveness of the proposed autofocusing algorithm for SAR image enhancement. In the second part of this paper the Modified Fractal Signature (MFS) method is presented. This method uses the “blanket” technique to provide useful information for SAR image classification. It is based on the calculation of the volume of a “blanket”, corresponding to the image to be classified, and then on the calculation of the corresponding fractal signature (MFS) of the image. We present here some results concerning the application of MFS method to the classification of SAR images. The MFS method is applied both in simulated data (comparison of a focused and an unfocused image) and in real-field data provided to us by SET 163 Working Group (comparison of a “town” area, “suburban” area and “sea” area). In these results it is clearly seen that the focusing of the SAR radar image clearly correlates with the value of MFS signature for the simulated data, and that the type of area can be distinguished by the value of MFS signature for the real data.