一种用于全息图自动重建的聚焦评价方法

A Focusing Criterion Function for Automatic Reconstruction of Digital Holograms

自动聚焦是数字全息技术中实现自动重建的关键技术之一,而自动聚焦的关键是寻找性能可靠的聚焦评价函数。通过对现有聚焦评价函数的深入研究分析,及对图像清晰度相关理论的学习,面向海洋生物全息图提出了一种新的聚焦评价函数,该聚焦评价函数的依据是局部标准差分布能够反映图像细节信息的特性及标准差自身具有较高稳健性的特征,以标准差作为离散程度度量,将全息图重建像细节的离散程度作为其清晰度指标,并将清晰度最大值对应的重建距离作为最佳重建距离,以此完成自动聚焦。实验结果表明:与其他聚焦评价函数相比,该方法具有较好的单峰性、较高的灵敏度和信噪比(SNR)及较好的无偏性等性能,且对于海洋生物全息图具有较强的实用性和通用性。因此,该方法可以满足海洋生物全息图自动重建的需求。

The autofocus is a key technique for automatic reconstruction in digital holography, and the performance of the focusing criterion function influences the effect of the autofocus greatly. After analyzing and researching existing focusing criterion functions and studying the theories about image definition, for the digital holograms of marine microorganism, a new focusing criterion function is proposed. The method proposed is based on features that local standard deviation distribution is able to reflect image detail information and standard deviation which is taken as measurement of discrete degree has high robustness. It makes the discrete degree of reconstruction image detail as image definition index, and makes reconstruction distance corresponding to maximum of the definition as the best distance to accomplish the autofocus. Simulation results illustrate that compared with the existing methods, the proposed method has the stronger single peak, higher sensibility and signal to noise ratio （SNR） and better unbiasedness. In addition, it has strong universal and practical for the digital holograms of marine microorganism. So it can satisfy the requirements of automatic reconstruction of the marine microorganism holograms.