基于幅相分离和动态粒子群算法的SAR图像属性散射中心参数估计

Parameter Estimation of Attributed Scattering Centers in SAR Images Based on Amplitude-Phase Separation and DPSO

针对SAR图像属性散射中心估计问题,提出了基于幅相分离和动态粒子群算法的参数估计方法。首先利用传统的图像域分析方法对单个属性散射中心进行解耦。然后,利用位置参数仅仅与相位信息有关的特性,采取了幅度、相位分离进行参数优化的方法。同时,对幅度相位分离后得到的较为简洁的参数表达形式,利用运算量更小的动态粒子群算法(DPSO),进一步提高了参数优化的效率和精度。基于仿真数据对提出方法进行了测试,结果验证了其有效性。

This paper proposes a parameter estimation algorithm of attributed scattering centers in SAR images based on amplitude-phase separation and DPSO.First,the traditional image domain analysis is used to separate a single attributed scattering center.Afterwards,the amplitude and phase of the scattering center are used for parameter estimation,respectively,considering that the position parameters are only related to the phase.Meanwhile,for the concise formulations of the amplitude and phase,the dynamic particle swarm optimization(DPSO)is employed to perform the parameter optimization,which can further improve the efficiency and precision of parameter estimation.The proposed method is evaluated based on the simulated data.And the results validate the effectiveness of the proposed method.