Research On 3D Point Cloud Reconstruction Method Of Circular SAR Typical Target Based On RD Back Projection

Circular Synthetic Aperture Radar(CSAR),as an all-round high-resolution 3D imaging mode,has the advantage of all-round observation.SAR 3D imaging of complex structural facilities is a hot and difficult problem in the field of SAR imaging.The existing SAR 3D imaging relies on multiple channels or multiple flights in the elevation direction,which has high requirements on the radar system or data acquisition;furthermore,the digital elevation model(DEM)is calculated using the stereo image pair principle The method is no longer applicable for strongly anisotropic targets.Therefore,this paper proposes a 3D point cloud reconstruction method for typical targets with Circular SAR based on RD back projection,which makes full use of the all-round observation,deconvolution masking and elevation blurring advantages of Circular SAR to obtain all-round 3D images of the area with unknown a priori information in just one flight,and is capable of 3D point cloud reconstruction for complex structural targets.The main research elements of this paper are as follows.(1)RD back projection model is given,and equidistant constant Doppler(RD)equation is used to calculate the coordinates of three-dimensional points.To solve the problem that the flight velocity vector of the radar sensor platform is offset due to the aircraft’s flight trajectory failing to meet the requirements of the standard circle and the unstable flight attitude.(2)To address the weak three-dimensional resolution of Circular SAR reconstruction using small-angle data and the inability to perform coherent accumulation when reconstructing using large-angle data because of the anisotropy of the target,the scattering coefficient varies with angle,non-coherent backward projection is used in this research.A combined coherent and incoherent method is proposed to first perform coherent imaging of small-angle subapertures to initially obtain a two-dimensional resolution,retain the magnitude information of the obtained images at each angle without using its phase information,perform incoherent accumulation of large angles in three-dimensional space,introduce target sparsity and sparse the target to improve the reconstruction effect.This method is suitable for sparse target reconstruction.Simulation and actual data experiments show that the reconstruction quality is high.(3)A three-dimensional reconstruction method for deblurring interferometric circumference SAR is proposed,which combines the advantages of high-precision altitude measurement of interferometry and all-round observation,unmasking and altitude blurring of circular SAR.The three-dimensional information acquisition ability of multi-azimuth data of circular SAR is used to solve the problem of target altitude blurring caused by interference phase blurring.One flight to acquire full scene full dimensional images.This method does not require target pre-modeling and 3D imaging grid construction,and is suitable for 3D fine imaging of complex structures in large scenes.The actual data experiment shows that the reconstruction quality is high.