论文部分内容阅读
基于孔径分解和图像递归融合的快速分解后向投影(FFBP)算法具备接近频域算法的运算复杂度和媲美后向投影(BP)算法的聚焦性能。但与频域成像算法不同,使用FFBP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,首先,提出了虚拟极坐标系作为FFBP算法的图像重建平面,为自聚焦方法的使用奠定了基础;其次,以基于回波数据的运动补偿为目标,充分利用FFBP算法多孔径递归融合的特点,将多孔径图像偏移(MAM)的相位估计方法嵌套到FFBP算法的各个阶段,从而实现MAM与FFBP算法的紧密相容;最后,通过实测数据处理验证了该方法的可行性和有效性。
The fast decomposition back projection (FFBP) algorithm based on aperture decomposition and image recursive fusion has the computational complexity of the near-domain algorithm and the focusing performance comparable to the back-projection (BP) algorithm. However, unlike the frequency domain imaging algorithm, the rectangular or polar coordinate system image reconstructed by FFBP algorithm can not meet the requirements of the traditional self-focusing method. In order to solve this problem, firstly, the virtual polar coordinate system is proposed as the image reconstruction plane of FFBP algorithm, which lays the foundation for the use of self-focusing method. Secondly, aiming at the motion compensation based on echo data, Recursive fusion, the method of phase estimation of multi-aperture image migration (MAM) is nested into each phase of FFBP algorithm to achieve the close compatibility between MAM and FFBP algorithm. Finally, the feasibility of this method is verified by the measured data processing Sexuality and effectiveness.