首页 > 论文 > 光学学报 > 40卷 > 21期(pp:2111001--1)

一种基于深度学习的光学合成孔径成像系统图像复原方法

Deep Learning Based Image Restoration Method of Optical Synthetic Aperture Imaging System

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

光学合成孔径成像系统中光学传递函数的频率响应下降,会不可避免地导致成像模糊,因此通常需要借助维纳滤波或盲解卷积算法来实现图像复原,最终获得清晰的高分辨率图像。提出一种基于U型卷积神经网络的深度学习框架,通过MATLAB软件构建数据集,以对网络进行训练,并将所训练的U型网络与盲解卷积算法的图像复原效果进行对比。数值仿真结果表明,在弱噪声条件下,U型网络在基于光学合成孔径成像系统的图像复原中展现出较强的复原能力以及一定的泛化能力和通用性,能够实现图像的快速盲复原,因而具有潜在的应用前景。

Abstract

The decrease of the frequency response of the optical transfer function in the optical synthetic aperture imaging system will inevitably lead to image blur. Therefore, it is usually necessary to use the Wiener filtering or blind deconvolution algorithm to achieve image restoration, and clear and high-resolution images are obtained finally. A deep learning frame based on a U-shaped convolutional neural network is proposed. The data set is constructed by the MATLAB software to train the network. The image restoration effects of the trained U-shaped network and blind deconvolution algorithm are compared. The numerical simulation results show that the U-shaped network has strong recovery ability, generalization ability, and versatility in the image restoration based on the optical synthetic aperture imaging system under the condition of weak noise. It can realize fast blind restoration for images and has potential application prospects.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:O436.1

DOI:10.3788/AOS202040.2111001

所属栏目:成像系统

基金项目:国家自然科学基金、国家自然科学基金委员会与中国工程物理研究院联合基金、陕西省自然科学基础研究计划、中央高校基本科研业务费专项资金;

收稿日期:2020-06-19

修改稿日期:2020-07-15

网络出版日期:2020-11-01

作者单位    点击查看

唐雎:西北工业大学物理科学与技术学院, 陕西 西安 710129陕西省光信息技术重点实验室, 陕西 西安 710129超常条件材料物理与化学教育部重点实验室, 陕西 西安 710129
王凯强:西北工业大学物理科学与技术学院, 陕西 西安 710129陕西省光信息技术重点实验室, 陕西 西安 710129超常条件材料物理与化学教育部重点实验室, 陕西 西安 710129
张维:西北工业大学物理科学与技术学院, 陕西 西安 710129陕西省光信息技术重点实验室, 陕西 西安 710129超常条件材料物理与化学教育部重点实验室, 陕西 西安 710129
吴小龑:中国工程物理研究院流体物理研究所, 四川 绵阳 621900
刘国栋:中国工程物理研究院流体物理研究所, 四川 绵阳 621900
邸江磊:西北工业大学物理科学与技术学院, 陕西 西安 710129陕西省光信息技术重点实验室, 陕西 西安 710129超常条件材料物理与化学教育部重点实验室, 陕西 西安 710129
赵建林:西北工业大学物理科学与技术学院, 陕西 西安 710129陕西省光信息技术重点实验室, 陕西 西安 710129超常条件材料物理与化学教育部重点实验室, 陕西 西安 710129

联系人作者:邸江磊(jiangleidi@nwpu.edu.cn); 赵建林(jlzhao@nwpu.edu.cn);

备注:国家自然科学基金、国家自然科学基金委员会与中国工程物理研究院联合基金、陕西省自然科学基础研究计划、中央高校基本科研业务费专项资金;

【1】Barakat R. Dilute aperture diffraction imagery and object reconstruction [J]. Optical Engineering. 1990, 29(2): 131-139.

【2】Lü Y K, Wu Y H. Development and key technologies of synthetic aperture ladar imaging [J]. Laser & Optoelectronics Progress. 2017, 54(10): 100004.
吕亚昆, 吴彦鸿. 合成孔径激光雷达成像发展及关键技术 [J]. 激光与光电子学进展. 2017, 54(10): 100004.

【3】Li D Y, Wu J, Wan L, et al. Elementary imaging theory on space-borne synthetic aperture ladar [J]. Acta Optica Sinica. 2019, 39(7): 0728002.
李丹阳, 吴谨, 万磊, 等. 天基合成孔径激光雷达成像理论初步 [J]. 光学学报. 2019, 39(7): 0728002.

【4】Bell K D, Boucher R H, Vacek R, et al. Assessment of large aperture lightweight imaging concepts[C]//1996 IEEE Aerospace Applications Conference. Proceedings, February 10, 1996, Aspen, CO, USA. New York: , 1996, 187-203.

【5】Fienup J R, Griffith D K, Harrington L, et al. Comparison of reconstruction algorithms for images from sparse-aperture systems [J]. Proceedings of SPIE. 2002, 4792: 1-8.

【6】Wang D, Han J, Liu H, et al. Experimental study on imaging and image restoration of optical sparse aperture systems [J]. Optical Engineering. 2007, 46(10): 103201.

【7】Wu Q Y, Qian L, Shen W M. Image recovering for sparse-aperture systems [J]. Proceedings of SPIE. 2005, 5642: 478-486.

【8】Liu L, Jiang Y S, Wang C W. Noise analysis and image restoration for optical sparse aperture systems[C]//2008 International Workshop on Education Technology and Training & 2008 International Workshop on Geoscience and Remote Sensing, December 21-22, 2008, Shanghai, China. New York: , 2008, 353-356.

【9】Wei X F, Geng Z X, Cao L, et al. A novel imaging performance index and pupil optimization method for optical synthetic aperture system [J]. Acta Optica Sinica. 2014, 34(11): 1111001.
魏小峰, 耿则勋, 曹力, 等. 一种新的光学合成孔径系统成像性能指标与光瞳优化方法 [J]. 光学学报. 2014, 34(11): 1111001.

【10】Zhou C H, Wang Z L, Zhang S Q, et al. Large aperture diffraction limited optical synthetic aperture system intermediate frequency MTF compensation [J]. Acta Optica Sinica. 2018, 38(4): 0411005.
周程灏, 王治乐, 张树青, 等. 大孔径衍射受限光学合成孔径系统MTF中频补偿 [J]. 光学学报. 2018, 38(4): 0411005.

【11】Wang F, Jiang M Q, Qian C, et al. Residual attention network for image classification[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 21-26, 2017, Honolulu, HI, USA. New York: , 2017, 6450-6458.

【12】Sarikaya R, Hinton G E, Deoras A. Application of deep belief networks for natural language understanding [J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing. 2014, 22(4): 778-784.Sarikaya R, Hinton G E, Deoras A. Application of deep belief networks for natural language understanding [J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing. 2014, 22(4): 778-784.

【13】Rivenson Y, Koydemir H C, Wang H, et al. Deep learning enhanced mobile-phone microscopy [J]. ACS Photonics. 2018, 5(6): 2354-2364.

【14】Lyu M, Wang H, Li G, et al. Learning-based lensless imaging through optically thick scattering media [J]. Advanced Photonics. 2019, 1(3): 036002.

【15】Li Y Z, Xue Y J. -09-26)[2020-06-19] [EB/OL]. Tian L. Deep speckle correlation: a deep learning approach towards scalable imaging through scattering media. 2018, org/abs/1806: 04139.Li Y Z, Xue Y J. -09-26)[2020-06-19] [EB/OL]. Tian L. Deep speckle correlation: a deep learning approach towards scalable imaging through scattering media. 2018, org/abs/1806: 04139.

【16】Zhang K, Zuo W M, Chen Y J, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising [J]. IEEE Transactions on Image Processing. 2017, 26(7): 3142-3155.

【17】Yan K T, Yu Y J, Huang C T, et al. Fringe pattern denoising based on deep learning [J]. Optics Communications. 2019, 437: 148-152.

【18】Wang K Q, Dou J Z, Qian K M, et al. Y-Net: a one-to-two deep learning framework for digital holographic reconstruction [J]. Optics Letters. 2019, 44(19): 4765-4768.

【19】Wang H, Lyu M, Chen N, et al. -01-10)[2020-06-19] . 2018.Wang H, Lyu M, Chen N, et al. -01-10)[2020-06-19] . 2018.

【20】Wang K Q, Li Y, Qian K M, et al. One-step robust deep learning phase unwrapping [J]. Optics Express. 2019, 27(10): 15100-15115.

【21】Wang K Q, Di J L, Li Y, et al. Transport of intensity equation from a single intensity image via deep learning [J]. Optics and Lasers in Engineering. 2020, 134: 106233.

【22】Wei X F, Geng Z X, Song X. Detection and removal of ringing artifact for optical synthetic aperture restoration image [J]. Optics and Precision Engineering. 2014, 22(11): 3091-3099.
魏小峰, 耿则勋, 宋向. 光学合成孔径复原图像的振铃探测与消除 [J]. 光学精密工程. 2014, 22(11): 3091-3099.

【23】Xu W H, Zhao M, Li H S. Non-iterative wavelet-based deconvolution for sparse aperture system [J]. Optics Communications. 2013, 295: 36-44.

【24】Kundur D, Hatzinakos D. Blind image deconvolution [J]. IEEE Signal Processing Magazine. 1996, 13(3): 43-64.

【25】Kundur D, Hatzinakos D. A novel blind deconvolution scheme for image restoration using recursive filtering [J]. IEEE Transactions on Signal Processing. 1998, 46(2): 375-390.

【26】Yang H L, Chiao Y H, Huang P H, et al. Blind image deblurring with modified Richardson-Lucy deconvolution for ringing artifact suppression . [C]// Proceedings of the 5th Pacific Rim conference on Advances in Image and Video Technology, November 20-23, 2011, Heidelberg: Springer. 2011, 240-251.

【27】Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation[M]. //Lecture Notes in Computer Science: , 2015, 234-241.

【28】Ioffe S. -02-11) [2020-06-19] [EB/OL]. Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. 2015, org/abs/1502: 03167.Ioffe S. -02-11) [2020-06-19] [EB/OL]. Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. 2015, org/abs/1502: 03167.

【29】Srivastava N, Hinton G E, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting [J]. Journal of Machine Learning Research. 2014, 15(1): 1929-1958.

【30】Horé A, Ziou D. Image quality metrics: , 2010, 2366-2369.

【31】Cheng G, Han J W, Lu X Q. Remote sensing image scene classification: benchmark and state of the art [J]. Proceedings of the IEEE. 2017, 105(10): 1865-1883.

引用该论文

Tang Ju,Wang Kaiqiang,Zhang Wei,Wu Xiaoyan,Liu Guodong,Di Jianglei,Zhao Jianlin. Deep Learning Based Image Restoration Method of Optical Synthetic Aperture Imaging System[J]. Acta Optica Sinica, 2020, 40(21): 2111001

唐雎,王凯强,张维,吴小龑,刘国栋,邸江磊,赵建林. 一种基于深度学习的光学合成孔径成像系统图像复原方法[J]. 光学学报, 2020, 40(21): 2111001

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF