在两步相移数字全息术中, 相移误差直接影响着零级像和共轭像的消除。为了消除相移误差, 获得实际相移值, 提出了一种基于相位统计特性的两步相移迭代算法。首先利用全息图的相位统计特性计算出初始相移值, 然后用数字全息图再现像抽样点强度偏差之和作为评价标准, 通过迭代计算寻找参考光的真正相移值。该算法能对任意未知相移量进行提取, 从而有效地消除零级像和共轭像, 提高了再现像质量。理论分析和光学实验结果证明了该方法的可行性和有效性。

In two-step phase-shifting digital holography, the elimination of the zero-order image and conjugate image is directly affected by the phase-shifting error. To eliminate the phase-shifting error and obtain the actual phase-shifting value, the two-step phase-shifting iterative algorithm is presented, based on the statistic property of the phase distribution of the diffraction field in recorded plane. In this method, the initial phase-shifting value is calculated directly from the holograms, and then the sum of the intensity bit errors between the reconstructed image and the holograms is taken as an evaluation function for an iterative algorithm to find the actual phase-shifting value. With this method, any unknown actual phase-shifting value can be found out, so the zero-order image and conjugate image are effectively eliminated, and the quality of the reconstructed image is enhanced. The feasibility and validity of this method are demonstrated by theoretical analysis and optical experimental result.