-
随着计算机处理能力的提升,以数据为驱动的机器学习发展极为迅速,其中深度学习因其强大的“逆问题”求解能力被国内外学者纷纷引入全息成像中来解决同轴全息产生的“孪生像”问题,深度学习技术是从大量的样本数据集中学习系统输入与输出之间复杂的高维度关联,包括系统存在的误差与不确定因素,因此往往能够实现非线性成像。目前,深度学习主要的基础模型有卷积神经网络(Convolutional Neural Networks, CNN)、循环神经网络(Recurrent Neural Network, RNN)、生成对抗式网络(Generative Adversarial Networks, GAN)等[66]。
2017年,美国加州大学的Rivenson等人[67]证明通过CNN训练相位恢复算法和全息图像的重建过程,能够快速有效消除孪生像的影响(见图25(a))。2018年,该课题组又提出一种能够同时执行自动聚焦和相位恢复的深度学习同轴全息图的重建方法[68],通过随机散焦的全息图像及其对应的焦内相位恢复图像对CNN进行训练,显著提高了视场的景深以及重建速度。同年,中国科学院上海光学精密机械研究所司徒国海等人[69]提出同轴全息图端到端的eHolo-Net深度学习重建技术,实现了从单张同轴全息图中重建物体的相位信息(见图25(b))。2019年,西北工业大学赵建林等人[70]提出了一种端到端重建的Y-Net结构,并且从单张全息图中同时恢复了物体的振幅和相位(如图25(c))。此外,基于深度学习的全息技术在显微成像方面也有研究,2019年,Ozcan等人[71]提出一种基于GAN的深度学习框架来实现全息系统的超分辨率成像,该方法突破了显微物镜数值孔径的限制,不需要任何额外的参数优化并且能够广泛应用于各种相干成像技术。
然而实际成像链路中不可避免地受到散射介质的影响,为了将基于深度学习的全息技术更广泛地应用到实际生产生活中,2020年,司徒国海等人[73]首次提出利用该技术实现透过散射介质成像,将eHoloNet神经网络结构用于短相干数字全息术成像中,实验光路如图26所示,通过研究不同散射介质对神经网络性能的影响,该技术实现了从单幅全息图中重建动态散射介质后的目标,推动了深度学习全息技术在散射成像中应用。
同年,Zhang等人[74]提出将生成对抗网络GAN用于学习散斑图和GS(Gerchberg-Saxton)算法计算得到的相位图之间的函数关系(如图27所示),该方法证明了深度学习网络通过学习散射介质的散射机理来实现透过散射介质成像(如图28所示),并且该方法能同时恢复图像的低频和高频信息,进一步突显出图像的细节,具有良好的泛化性能。虽然深度学习的引入为数字全息成像和穿透散射介质成像带来的新的可能,但在实际情况下,特别是散射成像应用中预先训练样本的获取是极难甚至是不可能的,小样本或样本泛化是将来深度学习实际应用的一大挑战。
为便于对以上几种技术的特点进行对比,在表1中进行了简单归纳,分析了它们在散射成像应用的几个关键方面的优劣势。
表 1 基于全息技术的散射成像方法对比
Table 1. Comparison of scattering imaging methods based on holography
Computational
complexityImaging through
dynamic mediaPenetration Field of view Coherent gating type ★ ★★★ ★ ★★☆ Phase conjugation type ★★ − ★ ★★ Correlation
type★★★ ★★ ★★★ ★ Synthetic wavelength type ★★ ★ ★★ ★★★ Deep learning type ★★★ ★★ ★★ ★★☆
Research and application progress of holography technology in scattering imaging (invited)
-
摘要: 散射是光学成像中普遍存在的现象,成像路径中存在的烟雾、水体、生物组织等散射介质导致光束发生随机散射效应,使得像面处目标信息以杂乱无章的散斑形式存在,如何应对散射介质对成像的限制是当前光学成像领域的研究热点。全息技术能够记录和重建物体全部信息,是获取和解译光场信息的有力工具之一。近年来,传统全息以及相关全息理论被推广应用至散射成像领域,取得了一系列突破性成果,文中主要介绍与归纳了散射成像领域中应用全息技术的理论原理、发展历史及最新进展,并展望其发展前景。Abstract: Scattering is an ordinary phenomenon in optical imaging. The imaging light is scattered by scattering media, such as smoke, water and biological tissues, existing in imaging path. The influence of scattering disturbs the imaging target information from orderly pattern to random speckles on imaging plane. How to break the limitation of scattering media is an essential problem in the field of optical imaging. Holographic technology can record and reconstruct all the information of objects, and play an important role in light field information obtaining and interpreting. In recent years, traditional holography and correlation holography theories have been applied to the field of scattering imaging, and a series of outstanding results have been achieved. This paper mainly introduces and summarizes the theoretical principles, development and latest achievement of holography technology in the field of scattering imaging. And the vista of holographic scattering imaging is also discussed.
-
图 2 浑浊水体中实现数字全息散射成像示意图。(a)实验装置图;(b) 实验结果:清水中(b1), (b2)和流动的牛奶(b3), (b4)中,精子重建的振幅和相位 [41]
Figure 2. Schematic diagram of digital holographic scattering imaging in turbid water. (a) Experimental setup; (b) Experimental results: Reconstruction of sperm amplitude and phase in clean water (b1), (b2) and flowing milk (b3), (b4) [41]
图 4 动态成像结果。(a) 在没有参考光束的情况下,直接在水中观察到的蚊子幼虫图像;(b) 在没有参考光束的情况下,直接在浑浊介质中观察到的蚊子幼虫图像;(c) 在浑浊介质中观察到的蚊子幼虫数字全息图;(d) 对应的角谱重建强度图像[42]
Figure 4. Dynamic imaging results. (a) Image of mosquito larvae observed directly in water without reference beam; (b) Image of mosquito larvae observed directly in opalescent medium without reference beam; (c) Digital hologram of mosquito larvae observed in turbid medium; (d) Corresponding angular spectrum reconstructed intensity image[42]
图 6 两散射介质层之间的光控制方法示意图。(a) 实验示意图;(b) 样本光束的共轭相位图;(c) 二次相位图;(d) 预计算的相位图;(e) 由(b)、(c)、(d)三部分叠加的相位图[44]
Figure 6. Schematic of the presented approach for light control between two turbid layers. (a) Schematic of experimental setup; (b) Conjugated phase map of the sample beam; (c) A quadratic phase map; (d) A pre-calculated phase map; (e) Phase diagram superimposed by (b), (c) and (d)[44]
图 8 同轴相移数字全息三维成像的重构结果。(a) 目标的结构示意图;(b) 计算得到的三维切片结果;(c) 聚焦于网格处重建图像;(d) 聚焦于玻璃珠处重建图像[45]
Figure 8. Recovered results of three-dimensional imaging based on in-line phase-shift digital holography. (a) Structure diagram of imaging target; (b) Reconstructed three dimensional slice results; Reconstructed image with a focus on (c) the grid and (d) the glass bead[45]
图 13 实验结果。(a) 原始对象;(b) 散斑强度模式的自协方差及其中心相关峰被阻塞;(c) 重构对象;(d) 原始对象;(e) 透生物组织散射的光的散斑强度;(f) 散斑强度模式的自协方差被其中心相关峰被阻塞;(g) 重构对象[53]
Figure 13. Experimental results. (a) Original object; (b) The autocovariance of the speckle intensity pattern and its central correlation peak are blocked; (c) Refactoring the object; (d) Original object; (e) Speckle intensity of light scattered through biological tissue; (f) The autocovariance of the speckle intensity pattern and its central correlation peak are blocked; (g) Refactoring objects[53]
图 15 实验结果。(a) 记录下来的物体V的散斑图案;(b) 恢复的同轴全息图;(c)和(d) 恢复的振幅和相位分布;(e)和(f) 恢复到GG1后各平面的振幅和相位分布[54]
Figure 15. Experimental results. (a) The recorded speckle pattern of the object V; (b) The recovered coaxial hologram; (c), (d) The recovered amplitude and phase distributions; (e), (f) Amplitude and phase distribution of each plane after restoration to GG1[54]
图 16 散射介质后目标相位成像实验装置。激光:氦氖激光器;MO:显微镜物镜;P:针孔;HWP:半波板;BS:分束器;M:镜子;SLM:空间光调制器;L1、L2、L3:晶状体;GG:磨砂玻璃;CCD: 电荷耦合器件[55]
Figure 16. Experimental setup for phase imaging of target behind scattering medium. Laser: He-Ne laser; MO: Microscope objective; P: Pinhole; HWP: Half-wave plate; BS: Beam splitter; M: Mirror; SLM: Spatial light modulator; L1, L2, L3: Lenses; GG: Ground glass; CCD: Charge coupled device[55]
图 19 不同标度尺下鬼衍射全息(上)与鬼衍射全息显微(下)的实验结果。(a)~(f) 1.15 mm;(g)~(j) 34.5 μm;(k) 23.0 μm;(l) 11.5 μm[56]
Figure 19. Experimental results of ghost diffraction holography (upper) and ghost diffraction holographic microscopy (down) for different scale bars. (a)-(f) 1.15 mm; (g)-(j) 34.5 μm; (k) 23.0 μm ; (l) 11.5 μm[56]
图 21 散射介质测量的实验结果。(a) 实验光路图;(b) 成像字符U,尺寸15 mm×20 mm;(c) 成像路径中使用的散射介质: 一个220目的毛玻璃和一个4 mm厚的乳白色塑料板,两者都放置在一个1 cm的方格图案上,以显示能见度的下降;(d)~(g) 通过毛玻璃的重建结果;(h)~(k) 通过乳白色塑料板测量的重建结果[34]
Figure 21. Experimental results for measurements through scatterering media. (a) Experimental optical path; (b) Imaged character U with dimensions 15 mm×20 mm; (c) Scatterers used in the imaging path: A 220 grit ground glass diffuser and a milky plastic acrylic plate of 4 mm thickness, both placed 1 cm over a checker pattern to demonstrate the decay in visibility; (d)-(g) Reconstructions of measurements taken through the ground glass diffuser; (h)-(k) Reconstructions of measurements taken through the milky acrylic plate[34]
图 22 合成脉冲全息术实验。(a) 目标,由两个字符组成,纵向间距为33 mm;(b)~(e) 仅使用单个合成波长重建字符;(f)~(i) 当合成波长的数量为23个时的重建效果;(f)和(g) 合成脉冲序列的脉冲距离[35]
Figure 22. Experimental demonstration of synthetic pulse holography. (a) Target, consisting of two characters with a longitudinal separation of 33 mm; (b)-(e) Reconstruction of the characters, using only NΛ=1 SWL; (f)-(i) Reconstruction results when the number of synthesized wavelengths is 23; The pulse distance of the synthesized pulse train can be seen in (f) and (g) [35]
图 23 非视域几何图形的实验设置。(a) 成像示意图;(b) NLoS实验光路图;(c) 粗糙目标表面;(d) 使用的目标图像:两个字符N和U,尺寸为15 mm×20 mm(加上黑色底座);(e) 用带透镜的光纤针注入参考光束,使光损失最小化[35]
Figure 23. Experimental setup for the NLoS geometry. (a) Imaging schematic; (b) Picture of the experimental NLoS setup; (c) Closeup image of the rough target surface; (d) Image of the used targets: Two characters N and U with dimensions 15 mm×20 mm (plus black mountings); (e) Injection of the reference beam with a lensed fiber needle for a minimized light loss[35]
表 1 基于全息技术的散射成像方法对比
Table 1. Comparison of scattering imaging methods based on holography
Computational
complexityImaging through
dynamic mediaPenetration Field of view Coherent gating type ★ ★★★ ★ ★★☆ Phase conjugation type ★★ − ★ ★★ Correlation
type★★★ ★★ ★★★ ★ Synthetic wavelength type ★★ ★ ★★ ★★★ Deep learning type ★★★ ★★ ★★ ★★☆ -
[1] Oliver B M. Sparkling spots and random diffraction [C]//Proceedings of the IEEE, 1963, 51(1): 220-221. [2] Feng S, Kane C, Lee P A, et al. Correlations and fluctuations of coherent wave transmission through disordered media [J]. Physical Review Letters, 1988, 61(7): 834-837. doi: 10.1103/PhysRevLett.61.834 [3] Freund I, Rosenbluh M, Feng S. Memory effects in propagation of optical waves through disordered media [J]. Physical Review Letters, 1988, 61(20): 2328. doi: 10.1103/PhysRevLett.61.2328 [4] Freund I. Looking through walls and around corners [J]. Physica A: Statistical Mechanics and its Applications, 1990, 168(1): 49-65. doi: 10.1016/0378-4371(90)90357-X [5] Feng S, Lee P A. Mesoscopic conductors and correlations in laser speckle patterns [J]. Science, 1991, 251(4994): 633-639. doi: 10.1126/science.251.4994.633 [6] Zuo Chao, Chen Qian. Computational optical imaging: An overview [J]. Infrared and Laser Engineering, 2022, 51(2): 20220110. (in Chinese) doi: 10.3788/IRLA20220110 [7] Shao X P, Su Y, Liu J P, et al. Connotation and system of computational imaging(Invited) [J]. Acta Photonica Sinica, 2021, 50(5): 0511001. (in Chinese) [8] Zhu L, Shao X P. Research progress on scattering imaging technology [J]. Acta Optica Sinica, 2020, 40(1): 011005. (in Chinese) [9] Cheng J. Theory of ghost scattering with incoherent light sources [J]. Physical Review A, 2016, 93(4): 043808. doi: 10.1103/PhysRevA.93.043808 [10] Cheng J. Theory of ghost scattering with biphoton states [J]. Photonics Research, 2017, 5(1): 41-45. doi: 10.1364/PRJ.5.000041 [11] Katz O, Heidmann P, Fink M, et al. Non-invasive single-shot imaging through scattering layers and around corners via speckle correlations [J]. Nature Photonics, 2014, 8(10): 784-790. doi: 10.1038/nphoton.2014.189 [12] Popoff S M, Lerosey G, Carminati R, et al. Measuring the transmission matrix in optics: an approach to the study and control of light propagation in disordered media [J]. Physical Review Letters, 2010, 104(10): 100601. doi: 10.1103/PhysRevLett.104.100601 [13] Popoff S M, Lerosey G, Fink M, et al. Controlling light through optical disordered media: transmission matrix approach [J]. New Journal of Physics, 2011, 13(12): 123021. doi: 10.1088/1367-2630/13/12/123021 [14] Yang H, Huang Y, Gong C, et al. Advances on techniques of breaking diffraction limitation using scattering medium [J]. Chinese Optics, 2014, 7(1): 1-25. (in Chinese) [15] Gabor D. A new microscopic principle [J]. Nature, 1948, 161(4098): 777-778. [16] Schnars U, Jüptner W P O. Digital recording and numerical reconstruction of holograms [J]. Measurement Science and Technology, 2002, 13(9): R85. doi: 10.1088/0957-0233/13/9/201 [17] Wang Y X, Wang D Y, Yang Y S. Application and analysis in the biomedicine field using digital holographic technology [J]. Chinese Journal of Lasers, 2014, 41(2): 0209002. (in Chinese) doi: 10.3788/CJL201441.0209002 [18] Wang H Y, Guo Z J, Zhang Z H. Image-plane digital holography for quantitative imaging of cells of chinese medical decoction pieces [J]. Chinese Journal of Lasers, 2012, 39(2): 0209002. (in Chinese) doi: 10.3788/CJL201239.0209002 [19] Javidi B, Nomura T. Securing information by use of digital holography [J]. Optics Letters, 2000, 25(1): 28-30. doi: 10.1364/OL.25.000028 [20] Lai S, Neifeld M A. Digital wavefront reconstruction and its application to image encryption [J]. Optics Communications, 2000, 178(4-6): 283-289. doi: 10.1016/S0030-4018(00)00641-6 [21] Van Heerden P J. Theory of optical information storage in solids [J]. Applied Optics, 1963, 2(4): 393-400. doi: 10.1364/AO.2.000393 [22] Leith E N, Kozma A, Upatnieks J, et al. Holographic data storage in three-dimensional media [J]. Applied Optics, 1966, 5(8): 1303-1311. doi: 10.1364/AO.5.001303 [23] Staebler D L, Amodei J J. Coupled-wave analysis of holographic storage in LiNbO3[M]//Landmark Papers on Photorefractive Nonlinear Optics. Washington, D C: World Scientific Publishing Company, 1995: 173-180. [24] Fan J Y, Yin B C, Wang W S. Three-dimensional deformation measurement based on double exposure digital holographic technology [J]. Infrared and Laser Engineering, 2014, 43(5): 1572-1576. (in Chinese) doi: 10.3969/j.issn.1007-2276.2014.05.041 [25] Leith E, Chen C, Chen H, et al. Imaging through scattering media with holography [J]. JOSA A, 1992, 9(7): 1148-1153. doi: 10.1364/JOSAA.9.001148 [26] Spitz E. Holographic reconstruction of objects through a diffusing medium in motion [J]. CR Acad Sci Ser B, 1967, 264: 1449-1452. [27] Zhang Y, Shin Y, Sung K, et al. 3D imaging of optically cleared tissue using a simplified CLARITY method and on-chip microscopy [J]. Science Advances, 2017, 3(8): e1700553. doi: 10.1126/sciadv.1700553 [28] Ghaneizad M, Kavehvash Z, Fathi H, et al. Incoherent holographic optical phase conjugation for imaging through a scattering medium [J]. IEEE Transactions on Instrumentation and Measurement, 2020, 70: 1-10. [29] Cui M, Yang C. Implementation of a digital optical phase conjugation system and its application to study the robustness of turbidity suppression by phase conjugation [J]. Optics Express, 2010, 18(4): 3444-3455. doi: 10.1364/OE.18.003444 [30] Naik D N, Ezawa T, Miyamoto Y, et al. Real-time coherence holography [J]. Optics Express, 2010, 18(13): 13782-13787. doi: 10.1364/OE.18.013782 [31] Takeda M, Wang W, Duan Z, et al. Coherence holography [J]. Optics Express, 2005, 13(23): 9629-9635. doi: 10.1364/OPEX.13.009629 [32] Naik D N, Singh R K, Ezawa T, et al. Photon correlation holography [J]. Optics Express, 2011, 19(2): 1408-1421. doi: 10.1364/OE.19.001408 [33] Somkuwar A S, Das B, Vinu R V, et al. Holographic imaging through a scattering layer using speckle interferometry [J]. JOSA A, 2017, 34(8): 1392-1399. doi: 10.1364/JOSAA.34.001392 [34] Willomitzer F, Rangarajan P V, Li F, et al. Synthetic wavelength holography: An extension of Gabor's holographic principle to imaging with scattered wavefronts [J]. arXiv preprint, 2019: 1912.11438. [35] Willomitzer F, Rangarajan P V, Li F, et al. Fast non-line-of-sight imaging with high-resolution and wide field of view using synthetic wavelength holography [J]. Nature Communications, 2021, 12(1): 1-11. doi: 10.1038/s41467-020-20314-w [36] Leith E N, Upatnieks J. Holographic imagery through diffusing media [J]. JOSA, 1966, 56(4): 523. doi: 10.1364/JOSA.56.000523 [37] Goodman J W, Huntley Jr W H, Jackson D W, et al. Wavefront-reconstruction imaging through random media [J]. Applied Physics Letters, 1966, 8(12): 311-313. doi: 10.1063/1.1754453 [38] Indebetouw G, Klysubun P. Imaging through scattering media with depth resolution by use of low-coherence gating in spatiotemporal digital holography [J]. Optics Letters, 2000, 25(4): 212-214. doi: 10.1364/OL.25.000212 [39] 陈忠斌. 生物芯片技术[M]. 北京: 化学工业出版社, 2005. [40] Pagonis D N, Nassiopoulou A G. Free-standing macroporous silicon membranes over a large cavity for filtering and lab-on-chip applications [J]. Microelectronic Engineering, 2006, 83(4-9): 1421-1425. doi: 10.1016/j.mee.2006.01.065 [41] Paturzo M, Finizio A, Memmolo P, et al. Microscopy imaging and quantitative phase contrast mapping in turbid microfluidic channels by digital holography [J]. Lab on a Chip, 2012, 12(17): 3073-3076. doi: 10.1039/c2lc40114b [42] Li S, Zhong J. Dynamic imaging through turbid media based on digital holography [J]. JOSA A, 2014, 31(3): 480-486. doi: 10.1364/JOSAA.31.000480 [43] Yaqoob Z, Psaltis D, Feld M S, et al. Optical phase conjugation for turbidity suppression in biological samples [J]. Nature Photonics, 2008, 2(2): 110-115. doi: 10.1038/nphoton.2007.297 [44] Qiao M, Liu H, Pang G, et al. Non-invasive three-dimension control of light between turbid layers using a surface quasi-point light source for precorrection [J]. Scientific Reports, 2017, 7(1): 1-8. doi: 10.1038/s41598-016-0028-x [45] Kodama S, Ohta M, Ikeda K, et al. Three-dimensional microscopic imaging through scattering media based on in-line phase-shift digital holography [J]. Applied Optics, 2019, 57(34): G345-G350. [46] Akkermans E, Montambaux G. Mesoscopic Physics of Electrons and Photons[M]. UK: Cambridge University Press, 2007: 35-45. [47] Goodman J W. Speckle Phenomena in Optics: Theory and Applications[M]. USA: Roberts and Company Publishers, 2007: 15-70. [48] Fienup J R. Reconstruction of an object from the modulus of its fourier transform [J]. Optics Letters, 1978, 3(1): 27-29. doi: 10.1364/OL.3.000027 [49] Kumar Singh R, Sharma M A. Recovery of complex valued objects from two-point intensity correlation measurement [J]. Applied Physics Letters, 2014, 104(11): 111108. doi: 10.1063/1.4869123 [50] Takeda M, Singh A K, Naik D N, et al. Holographic correloscopy-unconventional holographic techniques for imaging a three-dimensional object through an opaque diffuser or via a scattering wall: A review [J]. IEEE Transactions on Industrial Informatics, 2015, 12(4): 1631-1640. [51] Chen Z Y, Chen L, Fan W R. Progress on scattering imaging technologies based on correlation holography [J]. Laser & Optoelectronics Progress, 2021, 57(2): 0200001. (in Chinese) [52] Naik D N, Ezawa T, Miyamoto Y, et al. 3-D coherence holography using a modified Sagnac radial shearing interferometer with geometric phase shift [J]. Optics Express, 2009, 17(13): 10633-10641. doi: 10.1364/OE.17.010633 [53] Singh A K, Naik D N, Pedrini G, et al. Exploiting scattering media for exploring 3 D objects [J]. Light: Science & Applications, 2017, 6(2): e16219-e16219. [54] Vinu R V, Kyoohyun K, Somkuwar A S, et al. Imaging through scattering media using digital holography [J]. Optics Communications, 2019, 439: 218-223. doi: https://doi.org/10.1016/j.optcom.2019.01.080 [55] Chen L, Chen Z, Singh R K, et al. Imaging of polarimetric-phase object through scattering medium by phase shifting [J]. Optics Express, 2020, 28(6): 8145-8155. doi: 10.1364/OE.382551 [56] Vinu R V, Chen Z, Singh R K, et al. Ghost diffraction holographic microscopy [J]. Optica, 2020, 7(12): 1697-1704. doi: 10.1364/OPTICA.409886 [57] Wang Yonghong, Chen Weijie, Zhong Shimin, et al. Research progress in phase unwrapping technology and its application [J]. Measurement & Control Technology, 2018, 37(12): 1-7, 16. (in Chinese) doi: 10.19708/j.ckjs.2018.12.001 [58] Hayasaki Y, Tamano S, Yamamoto M, et al. Phase-shifting digital holography using two low-coherence light sources with different wavelength[C]//ICO20: Optical Information Processing. SPIE, 2006, 6027: 1247-1253. [59] Wojtkowski M. High-speed optical coherence tomography: Basics and applications [J]. Applied Optics, 2010, 49(16): D30-D61. doi: 10.1364/AO.49.000D30 [60] Andretzky P, Lindner M W, Herrmann J M, et al. Optical coherence tomography by spectral radar: dynamic range estimation and in-vivo measurements of skin[C]//Optical and Imaging Techniques for Biomonitoring IV. International Society for Optics and Photonics, 1999, 3567: 78-87. [61] Leith E N, Swanson G J. Achromatic interferometers for white light optical processing and holography [J]. Applied Optics, 1980, 19(4): 638-644. doi: 10.1364/AO.19.000638 [62] Dresel T, Häusler G, Venzke H. Three-dimensional sensing of rough surfaces by coherence radar [J]. Applied Optics, 1992, 31(7): 919-925. doi: 10.1364/AO.31.000919 [63] Willomitzer F, Li F, Balaji M M, et al. High resolution non-line-of-sight imaging with superheterodyne remote digital holography[C]//Computational Optical Sensing and Imaging. Optical Society of America, 2019: CM2 A. 2. [64] Ruffing B, Fleischer J. Spectral correlation of partially or fully developed speckle patterns generated by rough surfaces [J]. JOSA A, 1985, 2(10): 1637-1643. doi: 10.1364/JOSAA.2.001637 [65] Belmonte A. Statistical model for fading return signals in coherent lidars [J]. Applied Optics, 2010, 49(35): 6737-6748. doi: 10.1364/AO.49.006737 [66] Zhou Hongqiang, Huang Lingling, Wang Yongtian. Deep learning algorithm and its application in optics [J]. Infrared and Laser Engineering, 2019, 48(12): 1226004. (in Chinese) doi: 10.3788/IRLA201948.1226004 [67] Rivenson Y, Zhang Y, Günaydın H, et al. Phase recovery and holographic image reconstruction using deep learning in neural networks [J]. Light: Science & Applications, 2018, 7(2): 17141-17141. [68] Wu Y, Rivenson Y, Zhang Y, et al. Extended depth-of-field in holographic imaging using deep-learning-based autofocusing and phase recovery [J]. Optica, 2018, 5(6): 704-710. doi: 10.1364/OPTICA.5.000704 [69] Wang H, Lyu M, Situ G. EHoloNet: A learning-based end-to-end approach for in-line digital holographic reconstruction [J]. Optics Express, 2018, 26(18): 22603-22614. doi: 10.1364/OE.26.022603 [70] Wang K, Dou J, Kemao Q, et al. Y-Net: A one-to-two deep learning framework for digital holographic reconstruction [J]. Optics Letters, 2019, 44(19): 4765-4768. doi: 10.1364/OL.44.004765 [71] Liu T, De Haan K, Rivenson Y, et al. Deep learning-based super-resolution in coherent imaging systems [J]. Scientific Reports, 2019, 9(1): 1-13. doi: 10.1038/s41598-018-37186-2 [72] Wang Fei, Wang Hao, Bian Yaoming, et al. Applications of deep learning in computational imaging [J]. Acta Optica Sinica, 2020, 40(1): 0111002. (in Chinese) doi: 10.3788/AOS202040.0111002 [73] Dong S, Wang F, Wang H, et al. Learning-based short-coherence digital holographic imaging through scattering media[C]//AOPC 2020: Display Technology; Photonic MEMS, THz MEMS, and Metamaterials; and AI in Optics and Photonics. SPIE, 2020, 11565: 206-213. [74] Qianqian Z, Bo S, Chao W. Computational holographic imaging through scattering media using deep learning[C]//2020 IEEE International Conference on Artificial Intelligence and Computer Applications (ICAICA). IEEE, 2020: 258-260.