【摘要】 位移和相关(SAC)算法是一种自聚焦算法,在同样的估计精度下,SAC算法比其它算法的计算量至少要少一个数量级。但是,如果场景不均匀或存在强点,传统的SAC算法精度会下降。本文吸收了子孔径思想,改变了平均对象,降低了SAC算法对场景的要求。同时,新算法中还考虑了高次相位误差影响并加以克服以提高聚焦参数的精度。更多还原

【Abstract】 Shift - and-correlate (SAC) algorithm for autofocus with a computational complexity which is at least an order of magnitude lower than that of other alogrithms providing comparable accuracies. But if the scene is inhomogeneous. or there are dominant scatters, the accuracy of the SAC algorithm will decrease. This paper presents a novel autofocus algorithm which absorbes the means of subaperture. changes the objects averaged and ruduces the requirement of scens. At same time, the new algorithm considers the effect of higher -order phase errors and improves the focussing accuracy by removing them.更多还原