多普勒调频率是实现高分辨SAR成像的重要参数。机载SAR成像惯用的子孔径相关法(Map Drift,MD)是基于一个点目标回波数据模型提出的,与实际情况严重不符,且不适用于低对比度地域回波的参数估计,有时甚至完全失效。文中提出一种基于图像最小熵准则的多普勒调频率估计算法。该算法不基于任何回波数据模型,采用搜索方式实现最小熵准则解卷积,其突出优点是:鲁棒性强,其聚焦效果不依赖于回波数据,在无明显特征场景下尤为突出。实测数据的成像效果验证了该方法的可行性及有效性。

Doppler frequency modulation ratio is a key parameter for high resolution SAR imaging. The map drift algorithm, which is usually applied in airborne SAR imaging processing, is based on the model that there are echoes of only one point. The hypothesis heavily disaccords with the actual situation and is not applied to the parameter estimation for low contrast region, even sometimes fails. A novel Doppler frequency modulation ratio estimation algorithm based on the minimum entropy criteria is presented in this paper. The algorithm is not based on any echo model, which completes minimum entropy deconvolution by searching method. The outstanding merit of the algorithm is its high robustness and not relied on any echo data, especially for the scenes that there are not some obvious features. The imaging results of live data demonstrate the feasibility and validity of the method proposed in this paper