Sharpness-Based Autofocusing for Stripmap SAR Using an Adaptive-Order Polynomial Model

A novel autofocusing technique is developed for image from stripmap-mode synthetic aperture radar (SAR) data. The approach is based on maximizing the image sharpness function that induces the solution to maximum-posterior estimation. In this letter, closed-form expressions are derived for the gradients of the sharpness function with respect to the coefficients of the polynomial expansion, which makes the use of conjugate gradient algorithm available. Additionally, we also design a modified adaptive-order searching strategy, and it helps to remarkably reduce the computational load while maintaining the accuracy. Real airborne SAR data experiments and comparisons demonstrate the validity and effectiveness of the proposed algorithm.