Research On Technology And Application Of SAR Image Simulation

The simulation of SAR image is a way to study Synthetic Aperture Radar by simulating technology. It is very important in the process of studying Synthetic Aperture Radar, and following the development of SAR image application, it has gained its value in many aspects, such as designing and testifying SAR system in prophase, geometric correction and geocoding to SAR image, interpretation and target recognition to SAR image, testifying new algorithm in anaphase. Generally there are two ways to simulate SAR images, one is based on signal simulation and the other is based on character simulation. This paper detailedly describes the two simulating methods, and imports the facet model which is usually used in SAR signal simulation method into the other method. The shadow features in SAR images are studied. A new algorithm to define shadow regions is presented. Finally this paper studies the application of the SAR simulated image and presents the corresponding experiments. The main tasks accomplished are listed as follows:1. Analyzing the two methods of SAR image simulation in detail and presenting an improved simulating method based on character. The facet model is imported into this method, based on the facet models which are established on the digital elevation model, we can acquire more accurate parameters to calculate the gray of the simulated SAR image.2. The importance of the satellite's state vectors is analyzed, and two algorithms of fitting the satellite's state vectors are introduced, some experiments are presented.3. Studying shadow features of SAR image in depth. The features of single shadow regions and overlapping shadow regions are analyzed. This paper gives a new algorithm to define all shadow regions and experiments this algorithm, the results show that this algorithm is effective.4. Based on the experimental data of an ERS image, some simulating experiments are accomplished. The simulating results demonstrate that the textures of the simulated image are same with the real SAR image and perfectly reflect the special geometric distortion features of mountainous region in SAR image.5. The simulated and the real SAR image are matched automatically based on the SIFT algorithm, some homologous points are acquired, the application of positioning the ground control points in the real SAR image is accomplished. Finally, we use the check-point method to evaluate the precision of these ground control points.