The Image Focusing Quality Improvement Technology Of Spaceborne SAR Ship

Synthetic Aperture Radar(SAR)has the advantages of all-weather,all-day and wide-band observation.It is a new technical means for ocean monitoring and remote sensing.Ship target monitoring is an important direction for Spaceborne SAR marine applications which plays a key role in Military reconnaissance,fishery management,and transportation.Unlike airborne SAR,the Spaceborne SAR platform has fast moving speed and short synthetic aperture time,so ship target with weak maneuverable can still be focused in the SAR image in a local range comparable to the target size.This provides convenience for ship target detection and recognition.At present,the standard imaging processing idea of Spaceborne SAR is to design a distance-azimuth two-dimensional matched filter for cooperative targets with known motion parameters,and use FFT to realize the conversation from frequency domain to time domain.This method has wide adaptability and high processing efficiency,but there are two problems for ship target with unknown motion parameters: First,the motion of the ship makes its Doppler history different from that of the ground stationary target,which leads to azimuth filtering mismatch and azimuth defocusing;secondly,the Fourier transform imaging method is limited by Rayleigh resolution,and it is impossible to distinguish the scatters whose interval does not exceed the nominal resolution of the system.In order to improve the focusing quality of Spaceborne SAR ship target and provide a clear image with high-resolution for subsequent target feature extraction and recognition,these two problems are studied in this paper.Firstly,the characteristics of ship SAR images and their corresponding geometric structures are analyzed,and the characteristic that ship target images can be sparsely represented is pointed out.In the auto-focusing of the ship target SAR image,the geometric and signal models of the ship target of the Spaceborne SAR are established.The influence of the motion of ship target on the signal Doppler history is analyzed according to the signal model.Based on the classical algorithm and the space-variant characteristic of Doppler parameters,a space-variant auto-focusing algorithm is designed.The processing results of GF-3 satellite image verify the effectiveness of the algorithm.On the basis of Doppler phase compensation,the paper studies the super-resolution processing technology of ship target SAR image.An over-complete Fourier dictionary is constructed based on the sparse characteristic of the echo signal.The resolution of the image is improved effectively by convex optimization method.The validity of the algorithm is verified by one-dimensional simulation,two-dimensional simulation and the measured data of GF-3 satellite.