基于光场处理方法的太赫兹成像优化

辛涛; 彭烁; 董立泉; 张韶辉; 张存林

doi:10.37188/CO.EN.2022-0005

基于光场处理方法的太赫兹成像优化

辛涛^1,,
彭烁^2,,
董立泉²,
张韶辉^2, ,,
张存林^3, ,

1.
河北师范大学, 职业技术学院, 河北石家庄 050024
2.
北京理工大学, 光电学院, 北京 100081
3.
首都师范大学, 物理系, 北京 100048

详细信息

中图分类号: O441.4
计量
- 文章访问数: 549
- HTML全文浏览量: 387
- PDF下载量: 183
- 被引次数: 0
出版历程
- 收稿日期: 2022-03-31
- 修回日期: 2022-04-21
- 录用日期: 2022-05-13
- 网络出版日期: 2022-06-10

Improving terahertz imaging by light field processing

doi: 10.37188/CO.EN.2022-0005

XIN Tao^1
,,
PENG Shuo^2
,,
DONG Li-quan²,
ZHANG Shao-hui^{2
, ,},
ZHANG Cun-lin^{3
, ,}

1.
College of Career Technology, Hebei Normal University, Shijiazhuang 050024, China
2.
School of Optics and Photonics, Beijing Institute of Technology, Beijing 100081, China
3.
Department of Physics, Capital Normal University, Beijing 100048, China

Funds: Supported by Founding of National Key Research and Development Program of China (No. 2021YFC2202400); Foundation Enhancement Program (No. 2021-JCJQ-JJ-0823); National Natural Science Foundation of China (No. 61735003)

More Information

Author Bio:
XIN Tao (1971—), male, was born in Shi jia Zhuang, Hebei province. He received his bachelor's degree from Hebei University of Mechanical and Electrical Engineering in 1995 and his master's degree from Yanshan University in 2004. He now works at the Vocational and Technical College of Hebei Normal University. His research interests include terahertz imaging and computer graphics problems, particularly the optimization of terahertz images using light field techniques. E-mail: xintao71@mail.hebtu.edu.cn

PENG Shuo (1995—), female, was born in ShijiaZhuang. Hebei province. She received the B.Eng degree from Beijing Institute of Technology in 2017 and is currently working toward the D.Eng degree at School of Optics and Photonics, BIT. Her research interests include light-field imaging and 3D measurement. E-mail: pengshuoeve@163.com

Corresponding author: zhangshaohui@bit.edu.cn; cunlin_zhang@cnu.edu.cn

摘要

摘要:
近年来，太赫兹技术在检测、安保等诸多领域变得日益重要。如何扩大视场和提高成像质量是太赫兹成像的关键。针对以上问题，本文基于单相机扫描方式搭建了一个太赫兹光场成像系统以同时实现对太赫兹波空间和角度分布信息的利用。基于四维全光函数和双平面参数化方法，利用太赫兹光场传播过程中的强度一致性进行了重聚焦运算，通过对光场进行积分即可得到一系列对应不同视角、不同成像距离的结果。与初始的太赫兹成像结果相比，所搭建成像系统的视场和成像质量都得到了有效提升。在本文实验中，视场扩大了1.84倍，分辨率从1.3 mm提高到了0.7 mm。此外，部分被遮挡的目标信息也可通过使遮挡物离焦得到恢复。结果证明所搭系统能够提高太赫兹成像质量并拓展其功能，从而为太赫兹无损测量和安全检测提供了新的思路。
- 太赫兹成像 /
- 光场成像 /
- 太赫兹无损检测
Abstract:
Terahertz (THz) technology becomes increasingly important nowadays, especially in testing and security applications. Extending the field of view and increasing the imaging quality are both vital challenges for THz imaging. To address these problems, we build a THz light field imaging system based on a single-camera scanning configuration, which utilizes the 4D information of the spatial and angular distribution of THz waves. Based on the 4D plenoptic function and the parameterization method with two parallel planes, the intensity consistency of THz propagation is used for refocusing calculation, then a series of refocused images can be obtained by integrating original light field images corresponding to different imaging distances and views. Compared with the original THz imaging, the field of view and the imaging quality of the THz light field imaging are effectively improved. In our experiment, the field of view was enlarged by a factor of 1.84 and the resolution increased from 1.3 mm to 0.7 mm. Furthermore, information on some obscured targets could also be retrieved by defocusing the obstructions. This method could improve the imaging quality of THz imaging as well as expand its functions, which inspires a new way for THz nondestructive testing (NDT) and security inspection.
- terahertz imaging /
- light field imaging /
- THz nondestructive testing

HTML全文

Figure 1. Schematic diagram of the system. (a) The THz camera and the two-dimensional translation stage; (b) the light filter.

下载: 全尺寸图片幻灯片

Figure 2. Parametrization of the light field. (a) Pairs of points on two parallel planes; (b) parameterization of our system

下载: 全尺寸图片幻灯片

Figure 3. The resampling diagram of refocusing process. When changing the object distance, the spatial coordinate S of the reference sub-image u₀ are constant, while the corresponding object point moves from P to P_α. According to the geometric principles, we can calculate the new coordinate S_α of sub-image u, which is different from the coordinate S₁ corresponding to P

下载: 全尺寸图片幻灯片

Figure 4. A rubber ring in an envelope (grid size 9×3, D_u=D_v=10 mm). (a) The rubber ring (outer diameter is 25 mm; inner diameter is 17 mm) and the envelope; (b) the reference sub-image (u=4, v=2); (c) the refocused image (αL=720 mm); (d) normalized gradient image of the labelled regions of (b) and (c), respectively and their gradient values of labelled horizontal lines corresponding to t=197

下载: 全尺寸图片幻灯片

Figure 5. A steel board with hollowed parts in a plastic box (grid size 9×1, D_u=10 mm). (a) The steel board (length is 68 mm; width is 45 mm) and the plastic box; (b) the reference sub-image (u=5, v=1); (c) the refocused image (αL=730 mm); (d) intensity of labelled horizontal lines corresponding to t=174

下载: 全尺寸图片幻灯片

Figure 6. Metal wrenches fixed on a polyvinyl chloride(PVC) board by tape. (grid size 9×7, D_u=D_v=10 mm) (a) the wrenches (from left to right the diameters are 0.7 mm, 0.9 mm, 1.3 mm and 1.5 mm) and the PVC board; (b) the reference sub-image (u=3, v=4); (c) the refocused image (αL=680 mm); (d) the refocused image after histogram equalization.

下载: 全尺寸图片幻灯片

Figure 7. Metal wrenches (fixed camera and moving target, grid size 9×5, D_u=D_v=10 mm). (a) The wrenches and the PVC board; (b) the reference sub-image (u=5, v=3); (c) the refocused image (αL=700 mm); (d) the refocused image after histogram equalization

下载: 全尺寸图片幻灯片

Figure 8. Refocusing images of different sub-views (two PVC bottles with liquid fixed by tape in a high-density polyethylene plastic box, grid size 9×2, D_u=10 mm, D_v=20 mm). (a) The bottles (length is 50 mm; diameter is 13 mm; thickness is 2 mm) and the plastic box; (b) one primitive sub-image; (c)(d)(e)(f) are the refocused images with angular coordinates (4,1), (4,2), (1,1) and (9,1) respectively (αL=720 mm).

下载: 全尺寸图片幻灯片

Figure 9. Refocusing images of different depths. (a) The image focused on the bottles, αL=720 mm. (b) The image focused on the tape and the plastic box, αL=670 mm

下载: 全尺寸图片幻灯片

Figure 10. A steel holed cube blocked by a piece of wire mesh placed between it and the THz camera. (a) and (b) are two primitive sub-images showing the wire mesh and the obscured target respectively; (c) the refocused image

下载: 全尺寸图片幻灯片

Table 1. Coordinates of sample pixels in different sub-views

Sub-view (4,1) (4,2) (1,1) (9,1)

Pixel 1 (192,175) (192,138) (248,175) (99,175)
Pixel 2 (238,193) (238,156) (295,193) (145,193)

下载: 导出CSV

参考文献(24)

[1]	ZHANG Y D, DENG C, SUN W F, et al. Terahertz continuous-wave transmission imaging system and its application in security inspections[J]. Proceedings of SPIE, 2008, 6840: 684010.
[2]	MITTLEMAN D M. Twenty years of terahertz imaging [Invited][J]. Optics Express, 2018, 26(8): 9417-9431. doi: 10.1364/OE.26.009417
[3]	YAO J Q, LU Y, ZHANG B G, et al. New research progress of THz radiation[J]. Journal of Optoelectronics·Laser, 2005, 16(4): 503-510.
[4]	LEE A W, HU Q. Real-time, continuous-wave terahertz imaging by use of a microbolometer focal-plane array[J]. Optics Letters, 2005, 30(19): 2563-2565. doi: 10.1364/OL.30.002563
[5]	KARPOWICZ N, ZHONG H, ZHANG C L, et al. Compact continuous-wave subterahertz system for inspection applications[J]. Applied Physics Letters, 2005, 86(5): 054105. doi: 10.1063/1.1856701
[6]	VALUŠIS G, LISAUSKAS A, YUAN H, et al. Roadmap of terahertz imaging 2021[J]. Sensors, 2021, 21(12): 4092. doi: 10.3390/s21124092
[7]	YANG J, RUAN SH CH, ZHANG M, et al. Real-time continuous-wave imaging with a 1.63THz OPTL and a pyroelectric camera[J]. Optoelectronics Letters, 2008, 4(4): 295-298. doi: 10.1007/s11801-008-8036-0
[8]	XIAO H, ZHU F. Identification of dangerous goods in human THZ images[C]. Proceedings of the 2018 International Conference on Network, Communication, Computer Engineering (NCCE 2018), Atlantis Press, 2018: 899-903.
[9]	WOODWARD R M, WALLACE V P, ARNONE D D, et al. Terahertz pulsed imaging of skin cancer in the time and frequency domain[J]. Journal of Biological Physics, 2003, 29(2-3): 257-259.
[10]	TAO Y H, FITZGERALD A J, WALLACE V P. Non-contact, non-destructive testing in various industrial sectors with terahertz technology[J]. Sensors, 2020, 20(3): 712. doi: 10.3390/s20030712
[11]	WONG T M, KAHL M, BOLIVAR P H, et al. . Frequency Modulated Continuous Wave (FMCW) THz Image 3D Superresolution[Z]. arXiv: 1802.05457v1, 2018.
[12]	LEVOY M. Light fields and computational imaging[J]. Computer, 2006, 39(8): 46-55. doi: 10.1109/MC.2006.270
[13]	WU G CH, MASIA B, JARABO A, et al. Light field image processing: an overview[J]. IEEE Journal of Selected Topics in Signal Processing, 2017, 11(7): 926-954. doi: 10.1109/JSTSP.2017.2747126
[14]	FANG L, DAI Q H. Computational light field imaging[J]. Acta Optica Sinica, 2020, 40(1): 0111001. doi: 10.3788/AOS202040.0111001
[15]	YIN Y K, YU K, YU CH ZH, et al. 3D imaging using geometric light field: a review[J]. Chinese Journal of Lasers, 2021, 48(12): 1209001. (in Chinese)
[16]	LI Y, WANG X, ZHOU G, et al. Overview of 4D Light Field Representation[J]. Laser &Optoelectronics Progress, 2021, 58(18): 1811012. (in Chinese)
[17]	TIAN Y, ZENG H Q, HOU J H, et al. Light field image quality assessment via the light field coherence[J]. IEEE Transactions on Image Processing, 2020, 29: 7945-7956. doi: 10.1109/TIP.2020.3008856
[18]	COUILLAUD J, ZIOU D. Light field variational estimation using a light field formation model[J]. The Visual Computer, 2020, 36(2): 237-251. doi: 10.1007/s00371-018-1599-2
[19]	YAO T, SANG X ZH, WANG P, et al. Depth reconstruction for 3D light-field display based on axially distributed light field[J]. Optical Engineering, 2021, 60(5): 053103.
[20]	ZHOU G, LI S, LAM E Y. Light field image restoration in low-light environment[J]. Proceedings of SPIE, 2020, 11525: 115251H.
[21]	GUO C L, JIN J, HOU J H, et al. . Accurate light field depth estimation via an occlusion-aware network[C]. IEEE International Conference on Multimedia and Expo (ICME), IEEE, 2020.
[22]	JAIN R, GRZYB J, PFEIFFER U R. Terahertz light-field imaging[J]. IEEE Transactions on Terahertz Science and Technology, 2016, 6(5): 649-657.
[23]	LYU N F, ZUO J, ZHAO Y M, et al. Terahertz synthetic aperture imaging with a light field imaging system[J]. Electronics, 2020, 9(5): 830. doi: 10.3390/electronics9050830
[24]	LYU N F, ZUO J, ZHAO Y M, et al. Layer-resolving terahertz light-field imaging based on angular intensity filtering method[J]. Sensors, 2021, 21(22): 7451. doi: 10.3390/s21227451