Chinese Journal of Lasers, Volume. 51, Issue 11, 1101015(2024)

Advances in Optical Imaging in Scattering Media

Yaoming Bian1...2 and Guohai Situ1,23,* |Show fewer author(s)
Author Affiliations
  • 1Laboratory of Information Optics and Optoelectronic Technology, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201800, China
  • 2Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
  • 3School of Physics and Optoelectronic Engineering, Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, Hangzhou 310024, China
  • show less
    Figures & Tables(20)
    Principle of range-gated imaging[58]
    Underwater range gated imaging devices. (a)‒(c) LUCIE 1-3 series products developed by Canada[60-61]; (d) Aqua Lynx system developed by Sweden[62] ; (e) system developed by Beijing University of Science and Technology[64]; (f) system developed by the United States[63]; (g) “Phoenix Eyes” system developed by the Institute of Semiconductors, Chinese Academy of Sciences[65]
    Method based on statistical modeling of echoes[66]. (a) Diagram of the experimental setup; (b) experimental results, with the degree of scattering gradually increasing from left to right for both sets of data
    Application of ghost imaging in scattering environments. (a)(b) Experimental setup and results in the laboratory by Gong et al.[73]; (c)(d) experimental setup and results outdoors by Chen et al.[74]; (e)(f) experimental setup and results underwater by Li et al.[75]
    Principles of optical coherence tomography and its applications. (a) Schematic diagram of time-domain OCT; (b) schematic diagram of spectral-domain OCT; (c) schematic diagram of swept source OCT; (d) OCT prototype[98] ; (e1)(e2) retina of a normal human eye and its in vivo images by OCT[99]; (f1)(f2) photograph and schematic of an OCT endoscope for intraluminal imaging[100-101]; (g1)(g2) OCT images of normal colon and malignant adenoma[102]; (g3)(g4) tissue sections images corresponding to Figs. 5 (g1) and (g2)[102]
    Principle of wavefront shaping and experimental results[37,114]. (a) Experimental setup; (b) before wavefront shaping; (c) after wavefront shaping; (d) speckle patten before shaping; (e) refocused image after shaping; (f) conventional optical imaging system; (g) super-resolution imaging system based on wavefront shaping; (h) focal point of the conventional imaging; (i) speckle patten formed by a conventional imaging system through the scattering medium; (j) super-resolution refocused focal point by wavefront shaping; (k) optimized wavefront phase
    Principle and experimental results of optical phase conjugation[128] . (a) Conventional imaging system without scattering medium; (b) conventional imaging system with scattering medium; (c) optical phase conjugation imaging through scattering medium
    Principle and experimental results of transmission matrix measurements[38,134]. (a) Diagram of the experimental setup; (b1) speckle pattern through the scattering medium; (b2) single-point refocusing result; (b3) norm of the focusing operator; (b4) 3-point refocusing result; (c) reconstructed images; (d) correlation coefficient between reconstruction and object as a function of the asymmetric ratio
    Scattering imaging based on optical memory effect. (a) Principle of optical memory effect[42] ; (b) diagram of the experimental setup[12] ; (c) speckle pattern; (d) reconstructed object image
    Principle of NLOS imaging and experimental results. (a) Diagram of the TOF based NLOS imaging experimental setup[151]; (b) measured echoes; (c) recovered results; (d) diagram of the TOF based NLOS imaging experimental setup in the outfield[159]; (e) hidden human body model; (f) three-dimensional results of recovered model; (g) experimental setup of broad spectrum light NLOS imaging[140]; (h1) camera detected speckle pattern; (h2) recovered object image; (h3) original object; (i) white light NLOS imaging experimental setup[162] ; (j1) speckle pattern; (j2) optimized autocorrelation result; (j3) recovery result
    Histogram equalization. (a) Clear image, (b) fog image; (c) the histogram equalization result; (d)‒(f) grayscale histograms of Figs. 11 (a)‒(c)
    Schematic diagram of Retinex[173]
    Atmospheric scattering model[48]
    Dark channel prior dehazing process and results[53]. (a) Data process; (b) raw images; (c) recovery results; (d) estimated depth maps
    Method for maximizing image contrast[55]. (a) Clear image; (b) haze image; (c) distribution of edges of the region in the red rectangle; (d) input image; (e) recovery image; (f) estimated airlight
    Principles and results of polarization defogging[54]. (a) Relationship between captured light intensity and polarized angle; (b) brightest/worst polarized image; (c) darkest/best polarized image; (d) recovery results; (e) detail comparison of the best-polarized and recovery images
    Principle and results of angle-selection scattering imaging[210]. (a) Imaging schematic; (b) experimental setup; (c) single-frame SIR enhancement results, where I) is the raw image acquired by the imaging system without angle-selection, II) and III) with angle-selection, and IV), V), and VI) are the corresponding histogram equalization results, respectively; (d) schematic diagram of time-domain minimization filtering; (e) multi-frame recovery results for different visibility, where I) is the raw images, II), III) and IV) are the recovery results of single-frame, multi-frame superposition, and time-domain minimization filtering, respectively
    Deep learning based scattering imaging in active mode. (a) Experimental setup of static scattering medium[231] ; (b) recovery results of static scattering medium; (c) experimental setup of dynamic scattering medium[232] ; (d) recovery results of dynamic scattering medium
    Passive scattering imaging based on e-ink displays[259]. (a) Experimental Setup; (b) data sample, consisting of e-ink displays and real objects, used to train and test the network; (c) recovery results and comparison with conventional algorithms
    • Table 1. Summary of scattering imaging techniques

      View table

      Table 1. Summary of scattering imaging techniques

      Illuminationmode

      Imaging

      photons

      MethodAdvantageDisadvantage

      Activeartificial

      light

      source

      Utilizing

      scattered

      light

      Wavefront shaping

      Deep penetration,

      super‑resolution

      Need guide stars, slow measurement, static, invasive

      Optical phase

      conjugation

      Deeper penetration,

      DOPC is relatively fast

      Slow measurement, static
      Transmission matrix

      Deeper penetration,

      fast recovery after measurement

      Complex systems with high

      stability requirements, static

      Memory effectsNon‑invasive, system is relatively simpleField of view limited
      Non‑line‑of‑sightImaging around blocking obstaclesComplex system, low resolution

      Extracting

      ballistic

      light

      Range‑gated imaging

      Simple principle,

      high system maturity

      Penetration depth limited by multiple scattering
      Modeling of photons

      Good interpretability,

      resolution increase

      Limited ability to enhance

      penetration depth

      Optical coherence

      tomography

      Non‑invasive, high resolution

      Complex system, poor

      environmental adaptation

      Passivenatural

      sunlight

      source

      Image based

      enhancement

      Histogram

      equalization

      Significant contrast improvement,

      simple calculation

      Grayscale differences are

      corrupted, noise particle residue

      Retinex

      Good interpretability, better retention of

      grayscale differences

      Less effective with thick

      scattering

      Filtering

      Better recovery in thick

      scattering, simple calculation

      Prone to loss of high‑

      frequency detail

      Physical

      based

      recovery

      Dark channel

      prior

      Good interpretability,

      high color fidelity

      Few noise residues

      at dense fog

      Maximizing

      Contrast

      High contrast enhancementProne to color distortion

      Polarization

      difference

      More accurate scattered light

      based on multi‑frame estimation

      Requires multiple measurements, inability to overcome multiple scattering

      Angle

      selection

      Improve SIR from acquisition,

      enhance penetration depth, simple

      and universal system modifications

      Ability to enhance penetration depth diminishes as the field of view increases
      Deep learning based approaches

      Recovery details are improved

      over traditional methods

      Requires large amounts of data for training, poor generalization and interpretability
    Tools

    Get Citation

    Copy Citation Text

    Yaoming Bian, Guohai Situ. Advances in Optical Imaging in Scattering Media[J]. Chinese Journal of Lasers, 2024, 51(11): 1101015

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: laser devices and laser physics

    Received: Mar. 13, 2024

    Accepted: May. 7, 2024

    Published Online: Jun. 7, 2024

    The Author Email: Situ Guohai (ghsitu@siom.ac.cn)

    DOI:10.3788/CJL240678

    CSTR:32183.14.CJL240678

    Topics