Foshan University, School of Physics and Optoelectronic Engineering, Foshan, China.
Ji Hua Laboratory, Foshan, China.
J Biomed Opt. 2023 Mar;28(3):036006. doi: 10.1117/1.JBO.28.3.036006. Epub 2023 Mar 13.
Fourier ptychographic microscopy (FPM) enables quantitative phase imaging with a large field-of-view and high resolution by acquiring a series of low-resolution intensity images corresponding to different spatial frequencies stitched together in the Fourier domain. However, the presence of various aberrations in an imaging system can significantly degrade the quality of reconstruction results. The imaging performance and efficiency of the existing embedded optical pupil function recovery (EPRY-FPM) aberration correction algorithm are low due to the optimization strategy.
An aberration correction method (AA-P algorithm) based on an improved phase recovery strategy is proposed to improve the reconstruction image quality.
This algorithm uses adaptive modulation factors, which are added while updating iterations to optimize the spectral function and optical pupil function updates of the samples, respectively. The effectiveness of the proposed algorithm is verified through simulations and experiments using an open-source biological sample dataset.
Experimental results show that the proposed AA-P algorithm in an optical system with hybrid aberrations, recovered complex amplitude images with clearer contours and higher phase contrast. The image reconstruction quality was improved by 82.6% when compared with the EPRY-FPM algorithm.
The proposed AA-P algorithm can reconstruct better results with faster convergence, and the recovered optical pupil function can better characterize the aberration of the imaging system. Thus, our method is expected to reduce the strict requirements of wavefront aberration for the current FPM.
傅里叶叠层显微镜(FPM)通过获取一系列对应于不同空间频率的低分辨率强度图像,并在傅里叶域中拼接在一起,实现了具有大视场和高分辨率的定量相位成像。然而,成像系统中存在的各种像差会显著降低重建结果的质量。由于优化策略,现有的嵌入式光瞳函数恢复(EPRY-FPM)像差校正算法的成像性能和效率较低。
提出了一种基于改进相位恢复策略的像差校正方法(AA-P 算法),以提高重建图像质量。
该算法使用自适应调制因子,在更新迭代时添加,分别优化样本的光谱函数和光瞳函数更新。通过使用开源生物样本数据集进行模拟和实验,验证了所提出算法的有效性。
实验结果表明,与 EPRY-FPM 算法相比,该算法在具有混合像差的光学系统中能够恢复出轮廓更清晰、相位对比度更高的复杂幅度图像。图像重建质量提高了 82.6%。
所提出的 AA-P 算法可以更快地收敛并重建出更好的结果,并且恢复的光瞳函数可以更好地描述成像系统的像差。因此,我们的方法有望降低当前 FPM 对波前像差的严格要求。