Research On Scene Matching Technology Based On SAR Image

The vigorous development of scene matching technology has injected fresh blood into the research of intelligent precision guided weapons.With the increasing progress of SAR imaging resolution and computing power,scene matching technology based on SAR images has also ushered in new development opportunities.Scene matching technology includes the following three steps: pre-processing,scene matching and post-processing.In this paper,the preprocessing part mainly studies the speckle suppression algorithm for SAR images,the scene matching part mainly studies the scene matching algorithm based on SAR images,and the post-processing part mainly studies the sea surface ship target tracking algorithm based on SAR images.The dissertation content includes the following three aspects:In order to through difficulties that SAR image speckle affects the subsequent processing steps,this paper uses the NLM algorithm in the application of speckle suppression,and proposes a NLM speckle suppression algorithm for SAR images based on similar value calculation.First,combined with the NLM algorithm filtering schematic diagram,the model and idea of the NLM algorithm are described,and the method of using the integral graph to accelerate the NLM processing;then combined with the characteristics of the SAR image,it is analyzed step by step,and the wavelet transform is used for noise estimation,and the equivalent number of sights is used for adaptive determination.The size of the neighborhood window and the search window are combined with the pixel intensity difference,the structure difference and the spatial distance to calculate the similarity value between the SAR image blocks.The reason for using these processing steps is analyzed in principle.Finally,the actual SAR image is used for simulation experiments.The intuitive schematic diagram of the simulation results shows that the algorithm in this chapter retains more details while suppressing speckle noise.Quantitative analysis shows that using the algorithm in this chapter can improve the equivalent number of sights and edge retention index of SAR images,and effectively suppress speckle noise.In order to realize scene matching based on SAR image,SURF algorithm is used in scene matching application based on SAR image,and a SAR image scene matching algorithm based on SURF algorithm is proposed.First,it briefly introduces the SURF algorithm model,focusing on the feature point detection and description steps of the SURF algorithm;then,based on the SURF algorithm,it is analyzed step by step in combination with the simulation experiment,and the global comparison map is mainly added to the SAR image mask as the feature point detection.In the preprocessing step,the SURF descriptor and the Hu invariant moment feature are combined to describe the feature points,and the feature point matching scheme combining ratio detection and RANSAC method is adopted;finally,the simulation experiment is carried out by using the actual SAR image,and the experimental results show that the algorithm in this chapter can improve the accuracy and precision of matching between SAR images,and realize scene matching between SAR images.In order to realize the tracking of sea surface ships based on SAR images,a sea surface ship target tracking algorithm based on SAR images is summarized.First introduced the missileborne SAR observation coordinate system and image coordinate system,then analyzed the method of positioning the target in the real-time SAR image,and then studied the more commonly used target motion models in the application of maneuvering target tracking,mainly including CV model,The CA model and CS model,as well as the Kalman filter algorithm used for tracking filtering,and finally through the simulation of the ship's uniform acceleration linear motion,turning motion and the experimental results of missile proportional guidance,it shows that the Kalman filter method based on the CS model is effective for ship targets.The tracking effect is the best,and it provides technical support for the acquisition of guidance information based on the SAR image of the surface ship target.