基于深度学习的宽带消色差及大视场超构透镜设计

       超构透镜自提出以来被各领域作为传统透镜的微型替代而广泛研究，尤其是 在成像领域。但是囿于材料的有限折射率和制备工艺有限支持的深宽比，单元结 构所能提供的相位调控梯度也是有限的，从而超构透镜不得不在多个性能指标之 间折衷，这大大限制了成像系统的性能。通过将像差矫正推迟到成像后的图像处 理中，虽然可以突破上述限制，但是目前已有的方法大都是基于相位分布的优化， 一方面没有在设计过程中考虑加工时量化操作带来的影响，另一方面基于相位优 化的低优化自由度不足以优化出更高的性能。

      为解决上述问题，本文提出了一种基于结构参数优化的超构透镜成像系统设 计方法，端到端地联合优化超构透镜的单元结构参数和盲去卷积神经网络，使得 超构透镜能够生成高质量且各入射角度均衡的模糊图，然后去卷积神经网络从中 恢复出清晰图像。

       本文主要创新点如下：

     （1）根据到透镜中心的径向距离将超构透镜上所有单元分类，直接优化单元 结构参数从而提供更高的优化自由度，并在优化过程中加入量化操作使整个成像 系统适应加工时量化操作的影响，从而避免了设计与制造之间的性能损失。

     （2）在优化单元结构参数时，针对不同入射角度在损失函数中引入了相似性 正则项，使得不同入射角度的成像结果质量更加均衡，这有利于后续的去模糊神 经网络在每个角度都重建出清晰的图像，从而扩大了视场角。

     （3）基于已有的去模糊网络构建了适用于本成像系统的去模糊神经网络，并 实现了高质量的图像重建。

       根据所提出的超构透镜成像系统设计方法成功在单个大口径（0.5mm）超构 透镜上同时实现了宽波段消色差（390～780nm）和大视场（50^∘）成像。

[1] YU N F, GENEVET P, KATS M A, et al. Light propagation with phase discontinuities: Generalized laws of reflection and refraction[J/OL]. Science, 2011, 334: 333-337. DOI: 10.1126/science.1210713.
[2] SUNSL,YANGKY,WANGCM,etal. High-efficiency broadband anomalous reflection by gradient meta-surfaces[J/OL]. Nano letters, 2012, 12(12): 6223-6229. DOI: 10.1021/nl303266 8.
[3] LIN J, MUELLERJPB,WANGQ,etal. Polarization-controlled tunable directional coupling of surface plasmon polaritons[J/OL]. Science, 2013, 340(6130): 331-334. DOI: 10.1126/scie nce.1233746.
[4] ZHENGGX,MÜHLENBERNDH,KENNEYM,etal. Metasurfacehologramsreaching80% efficiency[J/OL]. Nature nanotechnology, 2015, 10(4): 308-312. DOI: 10.1038/nnano.2015.2.
[5] TARDIF C S, FAQIRI R, ZHAO H T, et al. Intelligent meta-imagers: From compressed to learned sensing[J/OL]. Applied Physics Reviews, 2022, 9(1): 11314. DOI: 10.1063/5.0076022.
[6] CHEN XZ, HUANGLL,MÜHLENBERNDH,et al. Dual-polarity plasmonic metalens for visible light[J/OL]. Nature Communications, 2012, 3(1): 1198. DOI: 10.1038/ncomms2207.
[7] ARBABI A, HORIE Y, BAGHERI M, et al. Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission[J/OL]. Na ture nanotechnology, 2014, 10(11): 937-943. DOI: 10.1038/nnano.2015.186.
[8] KHORASANINEJAD M, ZHU A Y, ROQUES-CARMES C, et al. Polarization-insensitive metalenses at visible wavelengths[J/OL]. Nano letters, 2016, 16(11): 7229-7234. DOI: 10.102 1/acs.nanolett.6b03626.
[9] KHORASANINEJADM,CHENWT,DEVLINRC,etal. Metalensesatvisiblewavelengths: Diffraction-limited focusing and subwavelength resolution imaging[J/OL]. Science, 2016, 352 (6290): 1190-1194. DOI: 10.1126/science.aaf6644.
[10] AVAYUO,ALMEIDAE,PRIORY,etal. Compositefunctionalmetasurfacesformultispectral achromatic optics[J/OL]. Nature Communications, 2016, 8(1): 14992. DOI: 10.1038/ncomms 14992.
[11] WANGSM,WUPC,SUVC,etal. Broadbandachromaticopticalmetasurfacedevices[J/OL]. Nature Communications, 2017, 8(1): 187. DOI: 10.1038/s41467-017-00166-7.
[12] CHENWT,ZHUAY,SANJEEVV,etal. Abroadbandachromatic metalens for focusing and imaging in the visible[J/OL]. Nature Nanotechnology, 2018, 13(3): 220-226. DOI: 10.1038/s4 1565-017-0034-6.
[13] FAN Z B, QIU H Y, ZHANG HL, et al. A broadband achromatic metalens array for integral imaging in the visible[J/OL]. Light, Science & Applications, 2019, 8(1): 67. DOI: 10.1038/s4 1377-019-0178-2.
[14] GROEVERB,CHENWT,CAPASSOF. Meta-lensdoubletinthe visible region[J/OL]. Nano letters, 2017, 17(8): 4902-4907. DOI: 10.1021/acs.nanolett.7b01888.
[15] SHALAGINOVMY,ANSS,YANGF,etal. Single-element diffraction-limited fisheye met alens[J/OL]. Nano letters, 2020, 20(10): 7429-7437. DOI: 10.1021/acs.nanolett.0c02783.
[16] LASSALLE E, MASS T W W, ESCHIMÈSE D, et al. Imaging properties of large field-of viewquadratic metalensesandtheirapplications to fingerprint detection[J/OL]. ACS Photonics, 2021, 8(5): 1457-1468. DOI: 10.1021/ACSPHOTONICS.1C00237.
[17] BARANIKOV A V, KHAIDAROV E, LASSALLE E, et al. Large field-of-view and multi color imaging with GaP quadratic metalenses[J/OL]. Laser & Photonics Reviews, 2024, 18(1): 2300553. DOI: 10.1002/lpor.202300553.
[18] LECUNY,BENGIOY,HINTONG. Deeplearning[J/OL]. Nature, 2015, 521(7553): 436-444. DOI: 10.1038/nature14539.
[19] LIU D J, TAN YX, KHORAME,etal. Training deep neural networks for the inverse design of nanophotonic structures[J/OL]. ACS Photonics, 2018, 5(4): 1365-1369. DOI: 10.1021/acsp hotonics.7b01377.
[20] LIU Z C, ZHU D Y, RODRIGUES S P, et al. Generative model for the inverse design of metasurfaces[J/OL]. Nano Letters, 2018, 18(10): 6570-6576. DOI: 10.1021/acs.nanolett.8b0 3171.
[21] ANSS, FOWLERC,ZHENGBW,etal. A deep learning approach for objective-driven all dielectric metasurface design[J/OL]. ACS Photonics, 2019, 6(12): 3196-3207. DOI: 10.1021/ acsphotonics.9b00966.
[22] AN S S, ZHENG B W, SHALAGINOV M Y, et al. Deep learning modeling approach for metasurfaces with high degrees of freedom[J/OL]. Optics Express, 2020, 28(21): 31932-31942. DOI: 10.1364/OE.401960.
[23] 安希鹏. 基于深度学习的宽谱消色差超透镜设计研究[D/OL]. 武汉: 华中科技大学,2021. DOI: 10.27157/d.cnki.ghzku.2021.003422.
[24] CHANGJL,WETZSTEING. Deepoptics for monocular depth estimation and 3D object de tection[C/OL]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). 2019: 10192-10201. DOI: 10.1109/ICCV.2019.01029.
[25] METZLER C A, IKOMA H, PENG Y F, et al. Deep optics for single-shot high-dynamic range imaging[C/OL]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recogni tion (CVPR). 2020: 1372-1382. DOI: 10.1109/CVPR42600.2020.00145.
[26] SITZMANNV,DIAMONDS,PENGYF,etal. End-to-end optimization of optics and image processing for achromatic extended depth of field and super-resolution imaging[J/OL]. ACM Transactions on Graphics (TOG), 2018, 37(4): 114:1-114:13. DOI: 10.1145/3197517.3201333.
[27] YU ZQ, ZHANGQB,TAOX,et al. High-performance full-color imaging system based on end-to-end joint optimization of computer-generated holography and metalens[J/OL]. Optics Express, 2022, 30(22): 40871-40883. DOI: 10.1364/OE.470419.
[28] CHANG J L, SITZMANN V, DUN X, et al. Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification[J/OL]. Scientific Reports, 2018, 8(1): 12324. DOI: 10.1038/s41598-018-30619-y.
[29] WU Y C, BOOMINATHAN V, CHEN H J, et al. PhaseCam3D —learning phase masks for passive single view depth estimation[C/OL]//2019 IEEE International Conference on Compu tational Photography (ICCP). 2019: 1-12. DOI: 10.1109/ICCPHOT.2019.8747330.
[30] LILL,SHUANGY,MAQ,etal. Intelligent metasurface imager and recognizer[J/OL]. Light, Science & Applications, 2019, 8(1): 97. DOI: 10.1038/s41377-019-0209-z.
[31] LI H Y, ZHAO H T, WEI M L, et al. Intelligent electromagnetic sensing with learnable data acquisition and processing[J/OL]. Patterns, 2020, 1(1). DOI: 10.1016/j.patter.2020.100006.
[32] DELHOUGNEP,IMANIMF,DIEBOLDAV,etal. Learnedintegratedsensingpipeline: Re configurable metasurface transceivers as trainable physical layer in an artificial neural network [J/OL]. Advanced Science, 2020, 7(3): 1901913. DOI: 10.1002/advs.201901913.
[33] TSENGE,COLBURNS,WHITEHEADJEM,etal. Neuralnano-opticsforhigh-quality thin lens imaging[J/OL]. Nature Communications, 2021, 12(1): 6493. DOI: 10.1038/s41467-021-26443-0.
[34] ZHU Z W, LIU Z C, ZHENG C X. Metalens enhanced ray optics: An end-to-end wave-ray co-optimization framework[J/OL]. Optics Express, 2023, 31(16): 26054-26068. DOI: 10.136 4/OE.496608.
[35] SUN J, CAO W, XU Z, et al. Learning a convolutional neural network for non-uniform mo tion blur removal[J/OL]. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015: 769-777. DOI: 10.1109/CVPR.2015.7298677.
[36] NAH S, KIM T H, LEE K M. Deep multi-scale convolutional neural network for dynamic scene deblurring[C/OL]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016: 257-265. DOI: 10.1109/CVPR.2017.35.
[37] TAOX,GAOHY,WANGY,etal. Scale-recurrentnetworkfordeepimagedeblurring[C/OL]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018: 8174-8182. DOI: 10.1109/CVPR.2018.00853.
[38] KUPYNO,BUDZANV,MYKHAILYCHM,etal.DeblurGAN:Blindmotiondeblurringusing conditional adversarial networks[C/OL]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018: 8183-8192. DOI: 10.1109/CVPR.2018.00854.
[39] KUPYN O, MARTYNIUK T, WU J R, et al. DeblurGAN-v2: Deblurring orders-of magnitude faster and better[C/OL]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). 2019: 8877-8886. DOI: 10.1109/ICCV.2019.00897.
[40] CAIJR,ZUOWM,ZHANGL. Darkandbrightchannelpriorembeddednetworkfordynamic scene deblurring[J/OL]. IEEE Transactions on Image Processing, 2020, 29: 6885-6897. DOI: 10.1109/TIP.2020.2995048.
[41] CHOSJ,JISW,HONGJP,etal.Rethinkingcoarse-to-fineapproachinsingleimagedeblurring [C/OL]//2021 IEEE/CVF International Conference on Computer Vision (ICCV). 2021: 4621 4630. DOI: 10.1109/ICCV48922.2021.00460.
[42] LIANGJY,CAOJ,SUNGL,etal. SwinIR:Imagerestorationusingswintransformer[C/OL]// 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). 2021: 1833-1844. DOI: 10.1109/ICCVW54120.2021.00210.
[43] LIUZ,LINYT,CAOY,etal. Swintransformer: Hierarchical vision transformer using shifted windows[C/OL]//2021IEEE/CVFInternationalConferenceonComputerVision(ICCV). 2021: 9992-10002. DOI: 10.1109/ICCV48922.2021.00986.
[44] WANGZD,CUNXD,BAOJM,etal. Uformer: Ageneral U-shaped transformer for image restoration[C/OL]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022: 17662-17672. DOI: 10.1109/CVPR52688.2022.01716.
[45] ZAMIR S W, ARORA A, KHAN S H, et al. Restormer: Efficient transformer for high resolution image restoration[C/OL]//2022 IEEE/CVF Conference on Computer Vision and Pat tern Recognition (CVPR). 2022: 5718-5729. DOI: 10.1109/CVPR52688.2022.00564.
[46] CHEN L Y, CHU X J, ZHANG X, et al. Simple baselines for image restoration[C/OL]// Computer Vision–ECCV 2022. 2022: 17–33. DOI: 10.1007/978-3-031-20071-7_2.
[47] SHRESTHA S, OVERVIG A C, LU M, et al. Broadband achromatic dielectric metalenses [J/OL]. Light: Science & Applications, 2018, 7(1): 85. DOI: 10.1038/s41377-018-0078-x.
[48] KHORASANINEJADM,SHIZJ,ZHUAY,etal.Achromaticmetalensover60nmbandwidth in the visible and metalens with reverse chromatic dispersion[J/OL]. Nano Letters, 2017, 17 (3): 1819-1824. DOI: 10.1021/acs.nanolett.6b05137.
[49] WANGSM,WUPC,SUVC,etal. Abroadband achromatic metalens in the visible[J/OL]. Nature Nanotechnology, 2018, 13(3): 227-232. DOI: 10.1038/s41565-017-0052-4.
[50] CHEN W T, ZHU A Y, SISLER J, et al. A broadband achromatic polarization-insensitive metalens consisting of anisotropic nanostructures[J/OL]. Nature Communications, 2018, 10 (1): 355. DOI: 10.1038/s41467-019-08305-y.
[51] WANG F, GENG G, WANG X, et al. Visible achromatic metalens design based on artificial neural network[J/OL]. Advanced Optical Materials, 2022, 10(3): 2101842. DOI: 10.1002/ad om.202101842.
[52] 王玉西. 介质型相位梯度超表面的研究及应用[EB/OL]. 武汉: 华中科技大学(2021). DOI: 10.27157/d.cnki.ghzku.2021.002219.
[53] WANGQ,ZHANGXQ,XUYH,etal. Broadbandmetasurface holograms: toward complete phase and amplitude engineering[J/OL]. Scientific Reports, 2016, 6(1): 32867. DOI: 10.1038/ srep32867.
[54] ZHOUF, LIU Y, CAI WP. Plasmonic holographic imaging with V-shaped nanoantenna array [J/OL]. Optics Express, 2013, 21(4): 4348-4354. DOI: 10.1364/OE.21.004348.
[55] PFEIFFER C, GRBIC A. Metamaterial Huygens’ surfaces: Tailoring wave fronts with reflec tionless sheets[J/OL]. Physical review letters, 2013, 110: 197401. DOI: 10.1103/PhysRevLet t.110.197401.
[56] CHENK,FENGYJ,MONTICONEF,etal.AreconfigurableactiveHuygens’metalens[J/OL]. Advanced Materials, 2017, 29(17): 1606422. DOI: 10.1002/adma.201606422.
[57] STAUDE I, MIROSHNICHENKO A E, DECKER M, et al. Tailoring directional scattering throughmagneticandelectricresonancesinsubwavelengthsiliconnanodisks[J/OL]. ACSnano, 2013, 7(9): 7824-7832. DOI: 10.1021/nn402736f.
[58] BERRY M V. Quantal phase factors accompanying adiabatic changes[J/OL]. Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences, 1984, 392(1802): 45-57. DOI: 10.1098/rspa.1984.0023.
[59] PANCHARATNAM S. Generalized theory of interference, and its applications[J/OL]. Pro ceedings of the Indian Academy of Sciences- Section A, 1956, 44: 247-262. DOI: 10.1007/bf 03046050.
[60] PU M B, LI X, MA X L, et al. Catenary optics for achromatic generation of perfect optical angular momentum[J/OL]. Science Advances, 2015, 1(9): e1500396. DOI: 10.1126/sciadv.1 500396.
[61] KHORASANINEJAD M, ZHU A Y, ROQUES-CARMES C, et al. Polarization-insensitive metalenses at visible wavelengths[J/OL]. Nano letters, 2016, 16(11): 7229-7234. DOI: 10.102 1/ACS.NANOLETT.6B03626.
[62] ARBABI A, HORIE Y, BALL A J, et al. Subwavelength-thick lenses with high numerical apertures and large efficiency based on high-contrast transmitarrays[J/OL]. Nature Communi cations, 2014, 6(1): 7069. DOI: 10.1038/ncomms8069.
[63] ARBABI A, ARBABI E, HORIE Y, et al. Planar metasurface retroreflector[J/OL]. Nature Photonics, 2017, 11(7): 415-420. DOI: 10.1038/nphoton.2017.96.
[64] LIANG Y, LIU H Z, WANG F Q, et al. High-efficiency, near-diffraction limited, dielectric metasurface lenses based on crystalline titanium dioxide at visible wavelengths[J/OL]. Nano materials, 2018, 8(5): 288. DOI: 10.3390/nano8050288.
[65] YARIVA,YEHP.Photonics: Opticalelectronicsinmoderncommunications[M]. USA:Oxford University Press, Inc., 2006.
[66] LUO X G, ZHANG F, PU M B, et al. Recent advances of wide-angle metalenses: Principle, design, and applications[J/OL]. Nanophotonics, 2022, 11(1): 1-20. DOI: 10.1515/nanoph-202 1-0583.
[67] KALVACHA,SZABÓZM.Aberration-freeflatlensdesignforawiderangeofincidentangles [J/OL]. Journal of The Optical Society of America B-optical Physics, 2016, 33(2): A66-A71. DOI: 10.1364/JOSAB.33.000A66.
[68] LINZ,GROEVERB,CAPASSOF,etal. Topology-optimized multilayered metaoptics[J/OL]. Physical review applied, 2018, 9: 044030. DOI: 10.1103/PhysRevApplied.9.044030.
[69] PU M, LI X, GUO Y, et al. Nanoapertures with ordered rotations: symmetry transformation and wide-angle flat lensing[J/OL]. Optics Express, 2017, 25(25): 31471-31477. DOI: 10.136 4/OE.25.031471.
[70] MARTINS A, LI K Z, LI J T, et al. On metalenses with arbitrarily wide field of view[J/OL]. ACSPhotonics, 2020, 7(8): 2073-2079. DOI: 10.1021/acsphotonics.0c00479.
[71] BURALLI DA, MORRISGM. Design of a wide field diffractive landscape lens[J/OL]. Ap plied optics, 1989, 28(18): 3950-3959. DOI: 10.1364/AO.28.003950.
[72] MUFFOLETTORP,TYLERJM,TOHLINEJE. ShiftedFresneldiffractionforcomputational holography[J/OL]. Optics Express, 2007, 15(9): 5631-5640. DOI: 10.1364/OE.15.005631.
[73] SYPEK M. Light propagation in the Fresnel region. New numerical approach[J/OL]. Optics Communications, 1995, 116(1): 43-48. DOI: 10.1016/0030-4018(95)00027-6.
[74] GOODMANJW. Introduction to Fourier optics[M]. 2005.
[75] MATSUSHIMA K, SHIMOBABA T. Band-limited angular spectrum method for numerical simulation of free-space propagation in far and near fields[J/OL]. Optics Express, 2009, 17 (22): 19662-19673. DOI: 10.1364/OE.17.019662.
[76] ZHANG W H, ZHANG H, JIN G F. Band-extended angular spectrum method for accurate diffraction calculation in a wide propagation range[J/OL]. Optics Letters, 2020, 45(6): 1543 1546. DOI: 10.1364/OL.385553.
[77] SHIMOBABAT,KAKUET,OKADAN,etal. Aliasing-reducedFresneldiffractionwith scale and shift operations[J/OL]. Journal of Optics, 2013, 15(7): 075405. DOI: 10.1088/2040-8978/ 15/7/075405.
[78] MATSUSHIMAK.Shiftedangularspectrummethodforoff-axisnumericalpropagation[J/OL]. Optics Express, 2010, 18(17): 18453-18463. DOI: 10.1364/OE.18.018453.
[79] MATSUSHIMAK. Formulationoftherotational transformation of wave fields and their appli cation to digital holography[J/OL]. Applied optics, 2008, 15(7): 075405. DOI: 10.1364/AO.4 7.00D110.
[80] NICOLASD,FINIZIOA,PIERATTINIG,etal. Angularspectrum method with correction of anamorphism for numerical reconstruction of digital holograms on tilted planes[J/OL]. Optics Express, 2005, 13(24): 9935-9940. DOI: 10.1364/OPEX.13.009935.
[81] MATSUSHIMAK. Computer-generated holograms for three-dimensional surface objects with shade and texture[J/OL]. Applied optics, 2005, 44(22): 4607-4614. DOI: 10.1364/AO.44.00 4607.
[82] 邱锡鹏. 神经网络与深度学习[M]. 北京: 机械工业出版社,2020.
[83] BOUVRIE J. Notes on convolutional neural networks[J]. Neural Nets, 2006, 1(1–8): 4607 4614.
[84] CHAKRABARTIA. Aneuralapproachtoblindmotion deblurring[C/OL]//Computer Vision ECCV2016: abs/1603.04771. 2016: 221-235. DOI: 10.1007/978-3-319-46487-9_14.
[85] KAUFMAN A, FATTAL R. Deblurring using analysis-synthesis networks pair[C/OL]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020: 5810 5819. DOI: 10.1109/CVPR42600.2020.00585.
[86] LI L RH, PANJS, LAI WS,et al. Dynamic scene deblurring by depth guided model[J/OL]. IEEE Transactions on Image Processing, 2020, 29: 5273-5288. DOI: 10.1109/TIP.2020.29801 73.
[87] SHEN ZY, LAI WH,XUTF,et al. Deep semantic face deblurring[C/OL]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018: 8260-8269. DOI: 10.1109/ CVPR.2018.00862.
[88] HEKM,ZHANGXY,RENSQ,etal. Deepresidual learning for image recognition[C/OL]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2016: 770-778. DOI: 10.1109/CVPR.2016.90.
[89] GAO H Y, TAO X, SHEN X Y, et al. Dynamic scene deblurring with parameter selective sharing and nested skip connections[C/OL]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2019: 3843-3851. DOI: 10.1109/CVPR.2019.00397.
[90] ZHANG K H, LUO W H, ZHONG Y R, et al. Deblurring by realistic blurring[C/OL]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020: 2734 2743. DOI: 10.1109/CVPR42600.2020.00281.
[91] RONNEBERGER O, FISCHER P, BROX T. U-Net: Convolutional networks for biomedical image segmentation[C/OL]//Medical Image Computing and Computer-Assisted Intervention MICCAI 2015. 2015: 234-241. DOI: 10.1007/978-3-319-24574-4_28.
[92] PANJS,BAIHR,TANGJH. Cascadeddeepvideodeblurringusingtemporal sharpness prior [C/OL]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2020: 3040-3048. DOI: 10.1109/CVPR42600.2020.00311.
[93] HUJ,SHENL,SUNG.Squeeze-and-excitationnetworks[C/OL]//2018IEEE/CVFConference on Computer Vision and Pattern Recognition. 2018: 7132-7141. DOI: 10.1109/CVPR.2018. 00745.
[94] HOU QB, ZHOU DQ, FENG J S. Coordinate attention for efficient mobile network design [C/OL]//2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021: 13708-13717. DOI: 10.1109/CVPR46437.2021.01350.
[95] WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[C/OL]// Computer Vision– ECCV 2018. 2018: 3-19. DOI: 10.1007/978-3-030-01234-2_1.
[96] JEGOU H, DOUZE M, SCHMID C. Hamming embedding and weak geometric consistency for large scale image search[C/OL]//Proceedings of the 10th EuropeanConferenceonComputer Vision: Part I. 2008: 304–317. DOI: 10.1007/978-3-540-88682-2_24.
[97] AGUSTSSON E, TIMOFTE R. NTIRE 2017 challenge on single image super-resolution: Dataset and study[C/OL]//2017 IEEE Conference on Computer Vision and Pattern Recogni tion Workshops (CVPRW). 2017: 1122-1131. DOI: 10.1109/CVPRW.2017.150.
[98] HOSSFELDT,HEEGAARDPE,VARELAM,etal. QoEbeyondtheMOS:anin-depth look at QoE via better metrics and their relation to MOS[J/OL]. Quality and User Experience, 2016, 1: 1-23. DOI: 10.1007/s41233-016-0002-1.
[99] WANG Z, BOVIK A, SHEIKH H, et al. Image quality assessment: from error visibility to structural similarity[J/OL]. IEEE Transactions on Image Processing, 2004, 13(4): 600-612. DOI: 10.1109/TIP.2003.819861.
[100] MITSA T, VARKUR K. Evaluation of contrast sensitivity functions for the formulation of quality measures incorporated in halftoning algorithms[C/OL]//1993 IEEE International Conference on Acoustics, Speech, and Signal Processing: Vol. 5. 1993: 301-304. DOI: 10.1109/ICASSP.1993.319807.
[101] WANG Z, SIMONCELLI E, BOVIK A. Multiscale structural similarity for image quality as sessment[C/OL]//The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers 2003: Vol. 2. 2003: 1398-1402. DOI: 10.1109/ACSSC.2003.1292216.
[102] SHEIKH H, BOVIK A, DE VECIANA G. An information fidelity criterion for image quality assessment using natural scene statistics[J/OL]. IEEE Transactions on Image Processing, 2005, 14(12): 2117-2128. DOI: 10.1109/TIP.2005.859389.
[103] DAMERA-VENKATA N, KITE T, GEISLER W, et al. Image quality assessment based on a degradation model[J/OL]. IEEE Transactions on Image Processing, 2000, 9(4): 636-650. DOI: 10.1109/83.841940.
[104] WANGZ,BOVIKA. Auniversalimage quality index[J/OL]. IEEE Signal Processing Letters, 2002, 9(3): 81-84. DOI: 10.1109/97.995823.
[105] SHEIKH H, BOVIK A. Image information and visual quality[J/OL]. IEEE Transactions on Image Processing, 2006, 15(2): 430-444. DOI: 10.1109/TIP.2005.859378.
[106] ZHANG R, ISOLA P, EFROS A A, et al. The unreasonable effectiveness of deep features as a perceptual metric[C/OL]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2018: 586-595. DOI: 10.1109/CVPR.2018.00068.
[107] SHEN Z Y, WANG W G, LU X K, et al. Human-aware motion deblurring[C/OL]//2019 IEEE/CVF International Conference on Computer Vision (ICCV). 2019: 5571-5580. DOI: 10.1109/ICCV.2019.00567.
[108] KETTUNEN M, HÄRKÖNEN E, LEHTINEN J. E-lpips: robust perceptual image similarity via random transformation ensembles[A]. 2019.
[109] MOORTHY A K, BOVIK A C. A two-step framework for constructing blind image quality indices[J/OL]. IEEE Signal Processing Letters, 2010, 17(5): 513-516. DOI: 10.1109/LSP.20 10.2043888.
[110] SAAD M A, BOVIK A C, CHARRIER C. Blind image quality assessment: A natural scene statistics approach in the DCT domain[J/OL]. IEEE Transactions on Image Processing, 2012, 21(8): 3339-3352. DOI: 10.1109/TIP.2012.2191563.
[111] MITTAL A, MOORTHY A K, BOVIK A C. No-Reference image quality assessment in the spatial domain[J/OL]. IEEE Transactions on ImageProcessing, 2012, 21(12): 4695-4708. DOI: 10.1109/TIP.2012.2214050.
[112] YE P, KUMAR J, KANG L, et al. Unsupervised feature learning framework for no-reference image quality assessment[C/OL]//2012 IEEE Conference on Computer Vision and Pattern Recognition. 2012: 1098-1105. DOI: 10.1109/CVPR.2012.6247789.
[113] MOORTHYAK,BOVIKAC. Blindimage quality assessment: From natural scene statistics to perceptual quality[J/OL]. IEEE Transactions on Image Processing, 2011, 20(12): 3350-3364. DOI: 10.1109/TIP.2011.2147325.
[114] MITTALA,SOUNDARARAJANR,BOVIKAC.Makinga“completelyblind”imagequality analyzer[J/OL]. IEEE Signal Processing Letters, 2013, 20(3): 209-212. DOI: 10.1109/LSP.20 12.2227726.
[115] LIU L X, LIU B, HUANG H, et al. No-reference image quality assessment based on spatial andspectral entropies[J/OL]. Signal Processing: Image Communication, 2014, 29(8): 856-863. DOI: 10.1016/j.image.2014.06.006.
[116] LI P, PRIETO L, MERY D, et al. Face recognition in low quality images: A survey[A]. 2018.
[117] PESTOURIE R, PéREZ-ARANCIBIA C, LIN Z, et al. Inverse design of large-area metasur faces[J/OL]. Optics Express, 2018, 26(26): 33732-33747. DOI: 10.1364/OE.26.033732.
[118] NOUREENS,MEHMOODMQ,ALIM,etal. Auniquephysics-inspireddeep-learning-based platform introducing a generalized tool for rapid optical-response prediction and parametric optimization for all-dielectric metasurfaces[J/OL]. Nanoscale, 2022, 14(44): 16436-16449. DOI: 10.1039/d2nr03644d.
[119] ALI J, AHMAD A, CHOI D Y. A metalens design for on- and off-center focusing with amor phous silicon hydrogenated (a-si:H)-based 1D array in visible spectrum[J/OL]. Electronics, 2023, 12(13). DOI: 10.3390/electronics12132953.
[120] KAMALI S M, ARBABI E, ARBABI A, et al. Highly tunable elastic dielectric metasurface lenses[J/OL]. Laser & Photonics Reviews, 2016, 10(6): 1002-1008. DOI: 10.1002/lpor.20160 0144.
[121] LI L G,WANGLZ,SONGWT,etal. Quantization-aware deep optics for diffractive snapshot hyperspectral imaging[C/OL]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022: 19748-19757. DOI: 10.1109/CVPR52688.2022.01916.
[122] LIU Z G, MATTINA M. Learning low-precision neural networks without straight-through estimator (STE)[C/OL]//Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence. 2019: 3066-3072. DOI: 10.24963/ijcai.2019/425.
[123] FOI A, TRIMECHE M, KATKOVNIKV,etal. Practical Poissonian-Gaussian noise modeling and fitting for single-image raw-data[J/OL]. IEEE Transactions on Image Processing, 2008, 17 (10): 1737-1754. DOI: 10.1109/TIP.2008.2001399.
[124] KINGMA D P, WELLING M. Auto-encoding variational bayes[J]. CoRR, 2013, abs/1312.6114.
[125] WILLIAMS R J. Simple statistical gradient-following algorithms for connectionist reinforce ment learning[J/OL]. Machine Learning, 2004, 8: 229-256. DOI: 10.1023/A:1022672621406.
[126] PUROHIT K, RAJAGOPALANAN. Motion deblurring with an adaptive network[A]. 2019.
[127] DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[A]. 2020.
[128] ZAMIR S W, ARORA A, KHAN S, et al. Multi-stage progressive image restoration[C/OL]// 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021: 14816-14826. DOI: 10.1109/CVPR46437.2021.01458.
[129] JOHNSON J, ALAHI A, Fei-Fei L. Perceptual losses for real-time style transfer and super resolution[C/OL]//LEIBE B, MATAS J, SEBE N, et al. Lecture Notes in Computer Science: Computer Vision– ECCV 2016. 2016: 694-711. DOI: 10.1007/978-3-319-46475-6_43.
[130] KINGMA DP,BAJ. Adam: Amethod for stochastic optimization[A]. 2014.