In remote-sensing applications, digital holography data often includes both phase errors from atmospheric turbulence and fully developed laser speckle from rough objects. When processing single-shot data, i.e., data from a single hologram, the high speckle contrast makes it more difficult to correct for atmospheric phase errors compared to scenarios where multiple speckle realizations are available for processing. A Bayesian phase-error correction algorithm [J. Opt. Soc. Am. A34, 1659 (2017)JOAOD60740-323210.1364/JOSAA.34.001659] was recently developed for use with single-shot data. The features of this approach are discussed and used to implement an alternative algorithm based on image-sharpness maximization. Algorithm performance is tested using simulated data for a range of signal-to-noise ratios (SNRs) and turbulence conditions. Using a combination of appropriate parameterization of the phase-error estimates and spatial binning for speckle-contrast reduction, the image-sharpness algorithm achieves performance comparable (better in the high-SNR regime but worse in the low-SNR regime) to the Bayesian approach. Limited experimental results are also presented.

中文翻译：

---

使用单次数据在数字全息术中进行相位误差校正。

在遥感应用中，数字全息数据通常既包括来自大气湍流的相位误差，也包括来自粗糙物体的充分发展的激光斑点。当处理单次数据（即来自单个全息图的数据）时，与多个斑点实现可用于处理的场景相比，高斑点对比度使校正大气相位误差更加困难。贝叶斯相位误差校正算法[J．选择。Soc。是。A34，1659（2017）JOAOD60740-323210.1364 / JOSAA.34.001659]最近被开发用于单次数据。讨论了此方法的功能，并将其用于实现基于图像清晰度最大化的替代算法。使用模拟数据针对一系列信噪比（SNR）和湍流条件测试算法性能。结合使用适当的相位误差估计值参数化和空间合并来减少斑点对比度，图像清晰度算法可实现与贝叶斯方法相当的性能（在高SNR情况下更好，而在低SNR情况下更差） 。还提供了有限的实验结果。