基于孔径分解和图像递归融合的快速分解后向投影算法具备接近频域算法的运算复杂度和媲美后向投影算法的聚焦性能。与频域成像算法不同,使用快速分解后向投影算法重建的直角坐标系图像或极坐标系图像均无法满足传统自聚焦方法的使用条件。为了解决这个问题,在原快速分解后向投影算法的基础上创新地引入了虚拟极坐标系,提出了改进快速分解后向投影算法,为自聚焦处理提供图像域与距离压缩相位历程域之间的傅里叶变换对关系。其次,建立全新的重叠子孔径构型,提出了嵌套多孔径相位梯度自聚焦的改进快速分解后向投影算法。最后,利用 X 波段聚束合成孔径雷达实测数据对该方法进行了验证。
Fast Factorized Back-Projection ( FFBP) is an effective algorithm constructed based on aperture factorization . It is shown that the image quality is barely sacrificed compared with Back-Projection ( BP) . Different from frequency domain algorithms , SAR images reconstructed by FFBP cannot satisfy the requirement of autofocus . To solve this problem , we propose an improved FFBP ( IFFBP) by introducing a fire-new coordinate system in this paper . Then , IFFBP can provide autofocus with Fourier transform pairs ( FTP) between the image domain and the range compressed phase history domain . With a different framework of the overlapped subaperture in different stages , phase gradient autofocus ( PGA ) can be embedded into IFFBP to obtain a robust estimation of the phase error . Finally , the processed results of the collected data set validate this approach in spotlight SAR imagery .