A dynamic scattering imaging device and method using the polarization characteristics of speckle field

The speckle image is collected by modulating linearly polarized light and polarization camera through the white light wide spectrum light source, combining independent component analysis and the Venasinchin theorem algorithm, and the imaging problem of dynamic scattering imaging under white light incoherent light is solved, achieving rapid and without first High-resolution imaging of information verification meets application needs in fields such as biomedicine.

CN115541536BActive Publication Date: 2024-11-05XIDIAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211024321.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-08-24
Publication Date
2024-11-05
Estimated Expiration
2042-08-24

AI Technical Summary

Technical Problem

The existing dynamic scattering imaging technology cannot achieve imaging with a spectral width of 100 nanometers under white light incoherent light, and requires a large amount of prior information or data, which cannot meet the application needs of biomedical, military security, and life sciences.

Method used

The white light wide spectrum light source is modulated into linearly polarized light, and a dynamic speckle light field is formed through a dynamic scattering medium. The polarization camera is used to collect speckle images of different polarization azimuth angles, and process them through independent component analysis and the Vinasinchin theorem and other algorithms. , achieve reconstruction of imaging goals.

Benefits of technology

It realizes rapid imaging of dynamic scattering media under white light source illumination, without reference to point sources or prior information, improves the time resolution and application range of imaging, can effectively remove background noise and separate clear target information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115541536B_ABST
    Figure CN115541536B_ABST
Patent Text Reader

Abstract

The present invention discloses a dynamic scattering imaging device and method using the polarization characteristics of a speckle field. The device comprises: a light source and an adjustment module, comprising a white light broadband light source and an adjustment module, wherein the adjustment module is used to adjust the light emitted by the white light broadband light source and modulate the adjusted light into linearly polarized light; an imaging module, comprising an imaging target located on a side of the adjustment module away from the white light broadband light source and a dynamic scattering medium located on a side of the imaging target away from the adjustment module; the dynamic scattering medium is used to form a dynamic speckle light field by causing multiple scattering of the linearly polarized light; an image detection and processing module, comprising a polarization camera located on a side of the dynamic scattering medium away from the imaging target and a processor located on a side of the polarization camera away from the dynamic scattering medium; the polarization camera is used to collect multiple speckle images with different polarization azimuths through a single exposure, and the processor is used to process the multiple speckle images to obtain an imaging image.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of scattering imaging and computational imaging, and in particular relates to a dynamic scattering imaging device and method using the polarization characteristics of a speckle field. Background Art

[0002] Dynamic scattering media refers to the disordered movement of particles inside the scattering medium over time. According to the number of scattering times that occur in dynamic scattering media, they can be divided into dynamic surface scattering media and dynamic volume scattering media. Surface scattering is a single scattering formed by a rough surface with random height. Its characteristic is that the scattering mainly comes from the surface in contact with the air. For example, the scattering on rotating frosted glass and rough metal surfaces is mainly surface scattering. Volume scattering is mainly caused by multiple scattering caused by scattering sub-particles that change with time inside the medium. Its characteristic is that the scattering of light after passing through the surface of the medium mainly occurs inside the medium, such as smoke, haze, turbid liquid, etc. In actual situations, dynamic surface scattering and volume scattering often exist at the same time. For example, the scattering caused by human skin and biological tissues includes both surface scattering and volume scattering. In the imaging process, the dynamic surface scattering imaging model can be equivalent to a simple linear random process, while the dynamic volume scattering process is accompanied by multiple interactions of particles inside the medium. The randomness can easily be annihilated by averaging, resulting in more information loss in the imaging results, making it more difficult to interpret and reconstruct. For static single scattering media, the speckle pattern formed does not change with time delay, which is reflected in the high correlation between speckles at different times, that is, there is no de-correlation effect between speckles; the speckle formed after passing through a dynamic multiple scattering medium will have obvious differences in pattern with the dynamic changes of the medium at the microscopic or macroscopic level, and the correlation between speckles at different times will gradually decrease, that is, dynamic speckle has a de-correlation effect, and its de-correlation time is extremely short. Therefore, the imaging method through static scattering media cannot solve the problem of dynamic multiple scattering imaging. However, there are still many problems in the existing dynamic scattering imaging methods that make it unable to meet the application requirements in biomedical imaging, military security, life sciences and other fields.

[0003] In the related art, there are devices and methods for imaging through dynamic scattering media, but these devices and methods either fail to achieve true imaging under white light incoherent light with a spectral width of hundreds of nanometers, or require a large amount of prior information or the collection of a large amount of data. Summary of the invention

[0004] In order to solve the above problems existing in the related art, the present invention provides a dynamic scattering imaging device and method using the polarization characteristics of the speckle field. The technical problem to be solved by the present invention is achieved through the following technical solutions:

[0005] The present invention provides a dynamic scattering imaging device using the polarization characteristics of a speckle field, comprising: a light source and adjustment module, an imaging module, and an image detection and processing module;

[0006] The light source and adjustment module include a white light broadband light source and an adjustment module located on one side of the white light broadband light source, wherein the adjustment module is used to adjust the light emitted by the white light broadband light source and modulate the adjusted light into linearly polarized light, wherein the light spot of the linearly polarized light covers the imaging target in the imaging module;

[0007] The imaging module comprises the imaging target located on a side of the adjustment module away from the white light broadband light source, and a dynamic scattering medium located on a side of the imaging target away from the adjustment module; the dynamic scattering medium is used to form a dynamic speckle light field by causing multiple scattering of the linearly polarized light;

[0008] The image detection and processing module comprises a polarization camera located on a side of the dynamic scattering medium away from the imaging target, and a processor located on a side of the polarization camera away from the dynamic scattering medium; the polarization camera is used to collect a plurality of speckle images with different polarization azimuths through a single exposure, and the processor is used to process the plurality of speckle images to obtain an imaging image of the imaging target.

[0009] The present invention further provides a dynamic scattering imaging method using the polarization characteristics of a speckle field, which is applied to the processor as described above, wherein the difference in polarization azimuth between two adjacent speckle images in the plurality of speckle images is a preset angle; the method comprises:

[0010] According to the plurality of speckle images, respectively calculating a first speckle image and a second speckle image;

[0011] Obtaining a speckle polarization difference image according to the first speckle image and the second speckle image;

[0012] Using the second speckle image and the speckle polarization difference image, and obtaining a polarization speckle image containing information of the imaging target through an independent component analysis algorithm;

[0013] Performing autocorrelation processing on the polarization speckle image containing the information of the imaging target to obtain autocorrelation information of the information of the imaging target;

[0014] According to the autocorrelation information, the Fourier amplitude of the imaging target is obtained by using the Wiener-Schintz theorem;

[0015] An imaging image of the imaging target is obtained by phase recovery and reconstruction according to the Fourier amplitude.

[0016] The present invention has the following beneficial technical effects:

[0017] The dynamic scattering imaging device comprises: a light source and adjustment module, an imaging module and an image detection and processing module; the light source and adjustment module comprises a white light broadband light source and an adjustment module located at one side of the white light broadband light source, the adjustment module is used to adjust the light emitted by the white light broadband light source, and modulate the adjusted light into linearly polarized light, the light spot of the linearly polarized light covers the imaging target in the imaging module; the imaging module comprises an imaging target located at a side of the adjustment module away from the white light broadband light source, and a dynamic scattering medium located at a side of the imaging target away from the adjustment module; the dynamic scattering medium is used to form a dynamic speckle light field by causing multiple scattering of the linearly polarized light; the image detection and processing module comprises a polarization camera located at a side of the dynamic scattering medium away from the imaging target, and a processor located at a side of the polarization camera away from the dynamic scattering medium; the polarization camera is used to collect multiple speckle images with different polarization azimuths through a single exposure, and the processor is used to process the multiple speckle images to obtain an imaging image of the imaging target. In this way, on the one hand, based on the polarization characteristics of white light broadband light source and dynamic speckle light field, imaging through dynamic scattering medium under white light source illumination can be realized. The imaging method does not require a reference point source, any prior information and narrow-band filters, and therefore has a wide range of applications. On the other hand, the polarization camera only needs to collect multiple speckle images with different polarization azimuths at the same time through a single exposure, and then the image of the imaging target can be reconstructed using the multiple speckle images collected simultaneously through a single exposure. The data acquisition and processing time is fast, which greatly improves the temporal resolution of imaging.

[0018] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic structural diagram of an exemplary dynamic scattering imaging device using the polarization characteristics of a speckle field provided in an embodiment of the present invention;

[0020] Figure 2 Another schematic structural diagram of an exemplary dynamic scattering imaging device using the polarization characteristics of a speckle field provided by an embodiment of the present invention;

[0021] Figure 3 An optional flow chart of a dynamic scattering imaging method using the polarization characteristics of a speckle field provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The present invention is further described in detail below with reference to specific embodiments, but the embodiments of the present invention are not limited thereto.

[0023] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0024] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification.

[0025] Although the present invention is described herein in conjunction with various embodiments, in the process of implementing the claimed invention, those skilled in the art may understand and implement other variations of the disclosed embodiments by viewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "one" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in the claims. Certain measures are recorded in different dependent claims, but this does not mean that these measures cannot be combined to produce good results.

[0026] There are three main types of existing methods for imaging through dynamic scattering media:

[0027] (1) Based on the idea of ​​extracting ballistic light imaging. In 2016, the team of Korean scientist Myungjin Cho applied photon counting technology to imaging through turbid water and dense fog. Through probability estimation theory, ballistic photons mixed in the scattering field were extracted to achieve three-dimensional imaging through dynamic turbid media. Only under short exposure conditions, the amount of data collected is large and the processing time is long. In 2018, Ramesh Raskar and others from the MIT Media Lab proposed the use of single-photon avalanche diode cameras combined with statistical estimation theory to select signal ballistic photons, and achieved imaging through dense fog with a thickness of 35 cm. In the same year, Peilong Hong from Ningbo University in China used incoherent and coherent two-photons to achieve amplitude and phase target imaging respectively. This method is assisted by a thin dynamic scattering layer and reaches the Heisenberg limit through constructive two-photon interference of modulated light. In 2019, Polish scientist Sudyka Julia used a single-pixel camera based on compressed sensing to achieve imaging through turbid water, but the amount of data collected was extremely large. Moreover, the number of ballistic photons decays exponentially with the increase of imaging distance, so the method based on ballistic light imaging cannot achieve long-distance imaging through dynamic scattering media, which limits its application in real life.

[0028] (2) The idea of ​​using scattered light for imaging. Due to the interference effect of scattered light, people regard the speckle pattern received by the detector as noise that hinders imaging. However, recovering target information from the speckle image is the essence of using scattered light for imaging, and it is also a hot topic studied by scholars in recent years. In 2016, E. Edrei and others from the University of Maryland used the shower curtain effect to estimate the autocorrelation information of the target by superimposing multiple frames of speckle illuminated target images and the Wiener-Khinchin theorem. Combined with the phase recovery algorithm, they realized imaging through high-speed dynamic surface scattering media. However, this method only targets dynamic surface scattering media, and has not studied dynamic volume scattering media, i.e., dynamic multiple scattering media. In addition, the amount of data collected is large and time-consuming, which means that it is impossible to image and track the target in real time, which limits its application. In 2017, Lipei Song's research group at Nankai University proposed a single-frame scattering imaging method based on the shower curtain effect through dynamic surface scattering media. This method combines holographic interference of the speckle field. Compared with the former, it only needs to collect single-frame speckle data, but this method requires prior information of the reference point source, and has not been extended to dynamic multiple scattering media, that is, this method cannot be applied to real-life scenarios. The above methods are only applicable to laser coherent light source illumination conditions, and incoherent light source illumination is not considered. In 2021, Dong Wang and others from Nanyang Technological University in Singapore used random optical scattering positioning technology to realize dynamic multiple scattering medium imaging through fresh eggshell membranes, with an imaging resolution of up to 100nm. However, this method still has the disadvantages of large amount of speckle image data required and low imaging time resolution. Although this method utilizes a red LED illumination source, a 10nm narrowband filter is added in front of the detector, which means that true spectral width imaging under white incoherent light of hundreds of nm is not achieved. The method using a narrowband filter does not solve the problem of low speckle contrast caused by the spectral width from the perspective of the physical model of imaging, and the practical adaptability of the filter remains to be studied.

[0029] (3) Image processing algorithms. The hottest deep learning algorithms in recent years can also achieve clear imaging of targets inside and outside dynamic volume scattering media. In 2019, Guihua Zeng's group at Shanghai Jiao Tong University used convolutional neural networks to reconstruct targets placed in a mixture of milk and water of different densities. In the same year, Wen Chen's group at the Hong Kong Polytechnic University reconstructed targets displayed on a digital micromirror device and hidden behind a fat emulsion solution based on cosine similarity and convolutional neural networks. In 2021, Guohai Situ's group at the Chinese Academy of Sciences used end-to-end deep neural networks to achieve imaging through a fat emulsion solution under green LED illumination. The disadvantage of these methods is that they require the acquisition of tens of thousands of frames of images to establish a large database as prior information, which greatly limits their application in real life.

[0030] In addition, there is currently a wide-spectrum scattering imaging method that uses the common-mode suppression of speckle field polarization. By collecting 72 speckle images formed through a static ground glass medium, the peak-to-correlation energy (PCE) of the autocorrelation function is used to find the speckle images corresponding to the maximum and minimum PCE values, that is, the images with the largest and smallest differences between the target information and the background noise. Based on these two images, the speckle polarization difference image is obtained, and the target is reconstructed using the speckle autocorrelation and phase recovery algorithms. However, the disadvantages of this technology are: a. It is only applicable to static single-scattering media, and the dynamic multiple-scattering media are not studied; b. Multiple exposures are required to collect 72 speckle images, and the imaging time resolution is low; c. It uses red light LED illumination with a spectral width of 16nm, which is not applicable to white light LED illumination.

[0031] Based on this, the inventors of this case discovered that if it is possible to utilize low-cost and easily available wide-spectrum illumination sources such as white light LEDs or natural light, without any prior or reference information, and to achieve rapid imaging through dynamic multiple scattering media after collecting speckle images through a single exposure, it will greatly promote the development of the scattering imaging field and improve the universality of scattering imaging technology.

[0032] Figure 1 FIG. 1 is an optional structural schematic diagram of a dynamic scattering imaging device using the polarization characteristics of a speckle field provided by an embodiment of the present invention. Figure 1 As shown, the device includes: a light source and adjustment module A, an imaging module B and an image detection and processing module C. Figure 1 As shown, the light source and adjustment module A includes a white light broadband light source 1 and an adjustment module 2 located on one side of the white light broadband light source. The adjustment module 2 is used to adjust the light emitted by the white light broadband light source 1 and modulate the adjusted light into linear polarized light. The spot of the linear polarized light covers the imaging target 3 in the imaging module B. The imaging module B includes an imaging target 3 located on the side of the adjustment module 2 away from the white light broadband light source 1 and a dynamic scattering medium 4 located on the side of the imaging target 3 away from the adjustment module 2; the dynamic scattering medium 4 is used to form a dynamic speckle light field by causing multiple scattering of the linear polarized light. The image detection and processing module C includes a polarization camera 5 located on the side of the dynamic scattering medium 4 away from the imaging target 3 and a processor 6 located on the side of the polarization camera 5 away from the dynamic scattering medium 4; the polarization camera 5 is used to collect multiple speckle images with different polarization azimuths through a single exposure, and the processor 6 is used to process the multiple speckle images to obtain an imaging image of the imaging target.

[0033] In the embodiment of the present invention, when the polarization camera 5 collects multiple speckle images in a single exposure, the light spot after passing through the dynamic multiple scattering medium is located at the center of the target surface of the polarization camera.

[0034] For example, a polarization camera can simultaneously receive speckle light field images with polarization azimuths of 0°, 45°, 90°, and 135° through a single exposure. These four images are randomly coded speckle images containing information of the imaging target and whose phases are modulated by the four polarization azimuths, respectively.

[0035] In some embodiments, Figure 2 As shown, the adjustment module 2 includes an aperture 21, a lens 22 and a polarizer 23; wherein the aperture 21 is located on one side of the white light broadband light source 1, the lens 22 is located on the side of the aperture 21 away from the white light broadband light source 1, and the polarizer 23 is located on the side of the lens 22 away from the aperture 2. The lens 22 can be a collimating lens, and the aperture 21 and the lens 22, these two devices together as a light path collimator, can better match the size of the light spot with the imaging target, and the aperture 21 and the lens 22 are aligned in the center so that the light spot can completely cover the imaging target. The polarizer 23 is a linear polarization modulation device, which is used to modulate the completely non-polarized light emitted by the light source into linearly polarized light.

[0036] In some embodiments, the spectrum width of the white light broadband light source is 200 nanometers to 400 nanometers.

[0037] In some embodiments, the dynamic scattering medium is at least any one of the following: onion skin, chicken breast tissue, egg shell membrane, turbid liquid, cloud and smoke particles.

[0038] Here, the onion skin, chicken breast tissue and egg shell membrane are all fresh onion skin, chicken breast tissue and egg shell membrane.

[0039] In some embodiments, the size of the imaging target is proportional to the distance between the dynamic multiple scattering medium and the distance between the dynamic multiple scattering medium and the polarization camera. For example, when the size of the imaging target 3 is 1.5 mm (millimeter), the dynamic multiple scattering medium can be located at 60 cm (centimeter) of the imaging target, and the polarization camera can be located 12 cm behind the dynamic multiple scattering medium.

[0040] In some embodiments, the dynamic multiple scattering medium used in the present invention is fresh onion epidermis with a thickness of 0.85 mm and fresh chicken breast tissue with a thickness of 1.2 mm. Due to the time-varying characteristics, the speckle patterns collected at different times change visibly with the naked eye, that is, the dynamic change of the speckle pattern, which is specifically reflected in the extremely low correlation between the speckle images at different times, and the fact that the fresh medium will gradually dry out. Therefore, the scattering medium is a dynamic medium. Because the optical thicknesses are 5.7 and 5.4 respectively, both parameters are greater than 1, indicating that the dynamic medium is a multiple scattering medium.

[0041] In some embodiments, the exposure time of the polarization camera can be adjusted according to the light source intensity and the medium thickness, for example, it can be 0.3 seconds.

[0042] The embodiments of the present invention can, on the one hand, realize imaging through a dynamic scattering medium under the illumination of a white light source based on the polarization characteristics of a white light broadband light source and a dynamic speckle light field. The imaging method does not require a reference point source, any prior information, and a narrow-band filter, and thus has a wide range of applications. On the other hand, the polarization camera only needs to simultaneously collect multiple speckle images with different polarization azimuths through a single exposure, and subsequently the image of the imaging target can be reconstructed using the multiple speckle images collected simultaneously through a single exposure. The data collection and processing time is fast, and the temporal resolution of the imaging is greatly improved.

[0043] The embodiment of the present invention further provides a dynamic scattering imaging method using the polarization characteristics of the speckle field, which can be applied to the processor 6 in the above-mentioned device; Figure 3 As shown, the method includes:

[0044] S101. Calculate a first speckle image and a second speckle image respectively according to a plurality of speckle images; a difference in polarization azimuth between two adjacent speckle images in the plurality of speckle images is a preset angle.

[0045] Exemplarily, the preset angle may be 45°.

[0046] Exemplarily, the number of the multiple speckle images can be set according to actual needs, for example, can be 4, etc., which is not limited in the embodiment of the present invention.

[0047] In the embodiment of the present invention, a plurality of Stokes parameters may be generated according to the plurality of speckle images, a first speckle image may be obtained by performing a first operation process on the plurality of Stokes parameters, and a second speckle image may be obtained by performing a second operation process on the plurality of Stokes parameters.

[0048] Here, the first speckle image I max , the second speckle image I min It can be expressed by formula (1) and (2) respectively:

[0049]

[0050]

[0051] In the above formulas (1) and (2), the information of the imaging target and the background noise are represented by O and B respectively, and the polarized light wave and the unpolarized light wave are represented by p and Unp respectively. According to formulas (1) and (2), we can know that the first speckle image I max Polarized light waves containing information about the imaged target Polarized light waves of background noise Unpolarized light waves that image the target's information and background noise The second speckle image includes polarized light waves of the imaging target information Polarized light waves of background noise Unpolarized light waves that image the target's information and background noise

[0052] S102: Obtain a speckle polarization difference image according to the first speckle image and the second speckle image.

[0053] In the embodiment of the present invention, the speckle polarization difference image may be obtained by calculating the pixel difference between corresponding pixels of the first speckle image and the second speckle image.

[0054] S103 . Using the second speckle image and the speckle polarization difference image, and using an independent component analysis algorithm, obtain a polarization speckle image containing information of the imaging target.

[0055] Here, since the correlation between the second speckle image and the speckle polarization difference image is smaller than the correlation between the first speckle image and the speckle polarization difference image, the information difference between the second speckle image and the speckle polarization difference image is the largest, which satisfies the prerequisite of the input parameters of the independent component analysis algorithm (the correlation between the two images is the smallest). Therefore, the second speckle image and the speckle polarization difference image can be used to obtain the polarization speckle image containing the information of the imaging target through the independent component analysis algorithm.

[0056] Here, the information of the imaging target and the background noise can be separated by the independent component analysis algorithm, and a polarization speckle image containing only the information of the imaging target can be calculated.

[0057] Here, the obtained polarization speckle image containing the information of the imaging target is a polarization speckle image containing only the information of the imaging target. The polarization speckle image has a high contrast, indicating that it only contains the information of the imaging target, while the background noise is suppressed.

[0058] S104 , performing autocorrelation processing on the polarization speckle image containing the information of the imaging target to obtain autocorrelation information of the information of the imaging target.

[0059] In the embodiment of the present invention, after the polarization speckle image containing only the information of the imaging target is obtained, an autocorrelation function may be used to perform autocorrelation processing on the polarization speckle image, thereby obtaining autocorrelation information of the information of the imaging target.

[0060] S105. According to the autocorrelation information, the Wiener-Schintz theorem is used to obtain the Fourier amplitude of the imaging target.

[0061] In the embodiment of the present invention, the process of obtaining the Fourier amplitude of the imaging target by using the Venatin-Chin theorem according to the autocorrelation information is as shown in formula (3):

[0062] F -1 {|F{O(x,y)}| 2}=O(x,y)☆O(x,y)≈I O (x,y)☆I O (x,y) (3)

[0063] Among them, |F{O(x,y)}| represents the Fourier amplitude of the imaging target, O(x,y) represents the information of the imaging target, and I O (x, y) represents the obtained polarization speckle image containing the information of the imaging target, and ☆ represents autocorrelation.

[0064] S106 , obtaining an imaging image of the imaging target by phase recovery and reconstruction according to the Fourier amplitude.

[0065] In the embodiment of the present invention, after obtaining the Fourier amplitude of the imaging target, the Fourier phase of the target can be obtained by a phase recovery algorithm. i2πO(ξ,η) ,Afterwards, the imaging image of the imaging target is reconstructed through inverse Fourier transform.

[0066] In the embodiments of the present invention, on the one hand, based on the polarization characteristics of a wide-spectrum white light source and a dynamic speckle light field, imaging through a dynamic scattering medium under illumination of a white light source can be realized. The imaging method does not require a reference point source, any prior information, and a narrow-band filter, and thus has a wide range of applications. On the other hand, a polarization camera only needs to simultaneously collect multiple speckle images with different polarization azimuths through a single exposure, and subsequently, the image of an imaging target can be reconstructed using the multiple speckle images collected simultaneously through a single exposure. The data acquisition and processing time is fast, and the temporal resolution of imaging is greatly improved. Moreover, multiple speckle images with different polarization azimuths can be combined with an independent component analysis algorithm to realize imaging through a dynamic random scattering medium under illumination of a wide-spectrum white light source, effectively removing the background noise effect caused by the spectrum width of the light source and multiple scattering of the dynamic medium, and being able to separate a polarization speckle image containing information of the imaging target, and using the autocorrelation information of the imaging target in the polarization speckle image and a phase recovery algorithm to restore a clear target.

[0067] In some embodiments, the plurality of speckle images include: a speckle image at a first polarization azimuth angle, a speckle image at a second polarization azimuth angle, a speckle image at a third polarization azimuth angle, and a speckle image at a fourth polarization azimuth angle; based on this, the above S101 can be implemented through S1011 to S1013:

[0068] S1011. Calculate a first Stokes parameter and a second Stokes parameter respectively according to the speckle image of the first polarization azimuth angle and the speckle image of the third polarization azimuth angle.

[0069] In the embodiment of the present invention, the sum of pixels between corresponding pixel positions of the speckle image of the first polarization azimuth angle and the speckle image of the third polarization azimuth angle may be calculated to obtain the first Stokes parameter;

[0070] Furthermore, the pixel difference between the corresponding pixel positions of the speckle image at the first polarization azimuth angle and the speckle image at the third polarization azimuth angle may be calculated to obtain the second Stokes parameter.

[0071] S1012: Calculate a third Stokes parameter according to the speckle image at the second polarization azimuth angle and the speckle image at the fourth polarization azimuth angle.

[0072] In the embodiment of the present invention, the pixel difference between the corresponding pixel positions of the speckle image at the second polarization azimuth angle and the speckle image at the fourth azimuth angle may be calculated to obtain the third Stokes parameter.

[0073] S1013, calculating a first speckle image and a second speckle image respectively according to the first Stokes parameter, the second Stokes parameter and the third Stokes parameter.

[0074] In the embodiment of the present invention, a first calculated value can be calculated according to the second Stokes parameter and the third Stokes parameter, and half of the sum of the first Stokes parameter and the first calculated value is used as the first speckle image; and half of the difference between the first Stokes parameter and the first calculated value is used as the second speckle image.

[0075] It should be noted that the first Stokes parameter, the second Stokes parameter and the third Stokes parameter are all obtained processed images; the calculations here are all for the calculation of the pixel values ​​between each corresponding pixel position in the image.

[0076] Exemplarily, the plurality of speckle images include: a speckle image with a polarization azimuth angle of 0°, a speckle image with a polarization azimuth angle of 45°, a speckle image with a polarization azimuth angle of 90°, and a speckle image with a polarization azimuth angle of 135°; based on this, the expressions of the first speckle image and the second speckle image are as shown in formulas (4) to (8):

[0077]

[0078]

[0079] I=I 0 +I 90 (6);

[0080] Q=I 0 -I 90 (7);

[0081] U 1 =I 45 -I 135 (8);

[0082] Among them, I max represents the first speckle image, I min represents the second speckle image, I represents the first Stokes parameter, Q represents the second Stokes parameter, U 1 represents the third Stokes parameter, I 0 represents the speckle image with a polarization azimuth of 0°, I 45 represents the speckle image with a polarization azimuth angle of 45°, I 90 represents the speckle image with a polarization azimuth angle of 90°, I 135 It shows the speckle image with a polarization azimuth angle of 135°.

[0083] In the embodiment of the present invention, the polarization camera only needs a single exposure to simultaneously collect speckle patterns at polarization azimuth angles of 0°, 45°, 90° and 135°, and the target can be reconstructed using four de-correlated dynamic speckle patterns. Compared with the existing through-dynamic scattering imaging method which requires hundreds of thousands of frames or more of speckle data, the imaging time resolution is greatly improved, making it possible to quickly interpret targets hidden behind media such as smoke, dust and haze.

[0084] In some embodiments, when the multiple speckle images include: a speckle image with a polarization azimuth angle of 0°, a speckle image with a polarization azimuth angle of 45°, a speckle image with a polarization azimuth angle of 90°, and a speckle image with a polarization azimuth angle of 135°, the speckle polarization difference image PDI may be expressed by the following formula (9):

[0085]

[0086] In some embodiments, when the first speckle image is represented by the above formula (1) and the second speckle image is represented by the above formula (2), it can be known from the physical meaning and mathematical expression of the first speckle image and the second speckle image that the proportion of non-polarized scattered light waves is small, so the maximum and minimum terms corresponding to the information of the imaging target and the background information are approximately equal, so the non-polarized scattered light parts in the first speckle image and the second speckle image can be eliminated by calculating the PDI image; thus, the mathematical and physical expression of the speckle polarization difference image PDI can be the following formula (10), which illustrates the maximum polarization term containing the information of the imaging target and the maximum polarization term of the background information in the figure, and the formula (10) is as follows:

[0087]

[0088] in, The maximum polarization term representing the information of the imaging target, The maximum polarization term representing the background information.

[0089] In some embodiments, the above S103 can be implemented through S1031 to S1041:

[0090] S1031. Use a matrix consisting of the second speckle image and the speckle polarization difference image as an input matrix.

[0091] For example, the input matrix can be expressed as X represents the input matrix; in the independent component analysis algorithm, the input matrix can also be expressed as X = A × S, where A represents the aliasing coefficient matrix and S represents the signal source matrix, so the input matrix can be further expressed as

[0092] Here, A and S are two unknown matrices.

[0093] S1032. Decompose the input matrix into a first matrix, a second matrix, and a noise coefficient matrix.

[0094] In the embodiment of the present invention, the first matrix consists of 0, 1 and -1; the second matrix consists of the first background speckle noise coefficient and the second background speckle noise coefficient; the third matrix consists of the polarization speckle image containing information of the imaging target and the polarization speckle image containing background information.

[0095] For example, the first matrix is The second matrix is The noise matrix is Among them, I O represents the polarization speckle image containing information of the imaging target, I B represents a polarized speckle image containing background information; represents the first background speckle noise coefficient, represents the second background speckle noise coefficient.

[0096] Here, I O ,I B , All are unknown elements; and All are I B The coefficient of .

[0097] Here, the process of decomposing the input matrix into the first matrix, the second matrix, and the noise coefficient matrix is ​​as follows:

[0098] because According to the above formula (9), we can get because The above formulas (1), (2) and (10) exist, and I max and I min yes The content of different components of

[0099] S1033. Perform singular value decomposition on the input matrix to obtain a first decomposition matrix, a second decomposition matrix, and a third decomposition matrix.

[0100] Exemplarily, the relationship between the input matrix and the first decomposition matrix, the second decomposition matrix and the third decomposition matrix can be expressed by the following formula (11):

[0101] X=U×D×V T =U×W×W -1 ×D×V T (11);

[0102] Among them, U represents the first decomposition matrix, D represents the second decomposition matrix, V T represents the third decomposition matrix, W represents the preset matrix, W -1 Represents the inverse matrix of the preset matrix.

[0103] Here, the first decomposition matrix, the second decomposition matrix, and the third decomposition matrix are matrices whose elements are known.

[0104] Here, each element in the preset matrix is ​​an unknown element, and the preset matrix is ​​determined by the first parameter, the second parameter, the third parameter and the fourth parameter. The preset matrix can be expressed as a represents the first parameter, α represents the second parameter, b represents the third parameter, and β represents the fourth parameter.

[0105] S1034. Obtain an aliasing coefficient matrix according to the first matrix and the noise coefficient matrix, and obtain a signal source matrix according to the second matrix.

[0106] Exemplarily, the aliasing coefficient matrix A and the signal source matrix S are respectively expressed by the following formulas (12) and (13):

[0107]

[0108]

[0109] S1035. According to the first decomposition matrix and the preset matrix, an aliasing coefficient matrix `4 is obtained, and according to the inverse matrix of the preset matrix, the second decomposition matrix and the third decomposition matrix, a signal source matrix is ​​obtained; wherein the preset matrix is ​​determined by the first parameter, the second parameter, the third parameter and the fourth parameter.

[0110] Exemplarily, the aliasing coefficient matrix A and the signal source matrix S are respectively expressed by the following formulas (14) and (15):

[0111] A = U × W (14);

[0112] S=W -1 ×D×V T (15);

[0113] S1036. Determine a first parameter and a second parameter according to the first decomposition matrix, the preset matrix, the first matrix, and the noise coefficient matrix.

[0114] Here, according to the above formulas (12) and (14), Thus, we know that because The first column of the resulting matrix is ​​known, and, Therefore, Since the first decomposition matrix U is known, the first parameter a and the second parameter α can be calculated based on it.

[0115] S1037. Determine a value set of the third parameter corresponding to the preset value set according to the preset matrix and the preset value set of the fourth parameter.

[0116] In the embodiment of the present invention, the preset value set of the fourth parameter may be [0, π].

[0117] Here, each time a value can be taken from [0,π] as the fourth parameter β, and the determinant of the matrix W is preset to 1, and a third parameter b corresponding to the fourth parameter β can be calculated. After that, a value can be taken from [0,π] again as the fourth parameter β, and the determinant of the matrix W is preset to 1, and a third parameter b corresponding to the fourth parameter β can be calculated, until all values ​​in [0,π] are taken, and the corresponding value set of the third parameter b is obtained.

[0118] S1038. Calculate a matrix set of preset matrices according to the first parameter, the second parameter, the preset value set and the value set.

[0119] In an embodiment of the present invention, each value in the preset value set of the first parameter a, the second parameter α, and the fourth parameter β, as well as each value in the value set of the third parameter b, can be substituted into W for calculation, so that a W matrix with the same number of values ​​as the value set of the third parameter b or the preset value set of the fourth parameter β can be obtained, thereby obtaining a matrix set of W.

[0120] S1039. Determine a plurality of groups of signals according to the matrix set, the second decomposition matrix, and the third decomposition matrix, each group of signals comprising: a polarization speckle image containing information of the imaging target, and a polarization speckle image containing background information.

[0121] In the embodiment of the present invention, when the matrix set of W is obtained, W can be obtained accordingly. -1 A matrix collection, which can be W -1 Each W in the matrix set -1 Substitute S = W -1 ×D×V T In calculating S, due to So, we can get -1 The same number of Each A set of signals.

[0122] S1040: Determine, from the multiple groups of signals, a group of signals having the smallest correlation between the polarization speckle image containing information of the imaging target and the polarization speckle image containing background information as the candidate signals.

[0123] In the embodiment of the present invention, for each of the multiple groups of signals obtained, I B with I O The correlation between them is calculated, and according to the calculation result, a group of signals is selected from multiple groups of signals as the candidate signals, and the I in the candidate signals is B with I O The correlation between them is minimal.

[0124] S1041. Use the polarization speckle image containing the information of the imaging target in the selected signal as the obtained polarization speckle image containing the information of the imaging target.

[0125] In the embodiment of the present invention, the I O The final polarization speckle image contains the information of the imaging target.

[0126] In general, the device and method provided by the present invention have the following beneficial technical effects:

[0127] 1. The device of the present invention is simple, highly stable and low in cost.

[0128] 2. The present invention utilizes the polarization characteristics of the dynamic speckle field, through the speckle patterns controlled by different polarization azimuth angles, combined with the Stokes parameter and independent component analysis algorithm, to achieve imaging through dynamic random scattering media under the illumination of a wide-spectrum white light source, effectively removes the background noise influence caused by the spectral width of the light source and multiple scattering of the dynamic medium, separates the target polarization speckle, and uses its autocorrelation information and phase recovery algorithm to restore a clear target.

[0129] 3. This method uses a polarization camera and only needs a single exposure to simultaneously collect 0°, 45°, 90° and 135° polarization azimuth speckle. That is, the target is reconstructed using four de-correlated dynamic speckle patterns. Compared with the existing through-dynamic scattering imaging method, which requires hundreds of thousands of frames or more of speckle data, it greatly improves the imaging time resolution, making it possible to quickly interpret targets hidden behind media such as smoke, dust and haze.

[0130] 4. Since the device and method use a wide-spectrum white light source for illumination, they lay a solid foundation for achieving imaging through clouds and fog under natural light.

[0131] The embodiment of the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and the computer program is used to execute some or all of the steps in the above-mentioned dynamic scattering imaging method using the polarization characteristics of the speckle field. The computer-readable storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0132] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A dynamic scattering imaging method using the polarization characteristics of a speckle field, characterized in that: The invention is applied to a processor in a dynamic scattering imaging device using the polarization characteristics of a speckle field, wherein the device comprises: a light source and adjustment module, an imaging module and an image detection and processing module; The light source and adjustment module include a white light broadband light source and an adjustment module located on one side of the white light broadband light source, wherein the adjustment module is used to adjust the light emitted by the white light broadband light source and modulate the adjusted light into linearly polarized light, wherein the light spot of the linearly polarized light covers the imaging target in the imaging module; The imaging module comprises the imaging target located on a side of the adjustment module away from the white light broadband light source, and a dynamic scattering medium located on a side of the imaging target away from the adjustment module; the dynamic scattering medium is used to form a dynamic speckle light field by causing multiple scattering of the linearly polarized light; The image detection and processing module comprises a polarization camera located at a side of the dynamic scattering medium away from the imaging target, and the processor located at a side of the polarization camera away from the dynamic scattering medium; the polarization camera is used to collect multiple speckle images with different polarization azimuths through a single exposure, and the processor is used to process the multiple speckle images to obtain an imaging image of the imaging target; the difference in polarization azimuth between two adjacent speckle images in the multiple speckle images is a preset angle; the multiple speckle images include: a speckle image with a first polarization azimuth, a speckle image with a second polarization azimuth, a speckle image with a third polarization azimuth, and a speckle image with a fourth polarization azimuth; The method comprises: calculating a first Stokes parameter and a second Stokes parameter respectively according to the speckle image of the first polarization azimuth angle and the speckle image of the third polarization azimuth angle; calculating a third Stokes parameter according to the speckle image at the second polarization azimuth angle and the speckle image at the fourth polarization azimuth angle; calculating a first speckle image and a second speckle image respectively according to the first Stokes parameter, the second Stokes parameter and the third Stokes parameter; Obtaining a speckle polarization difference image according to the first speckle image and the second speckle image; Using a matrix composed of the second speckle image and the speckle polarization difference image as an input matrix; Decomposing the input matrix into a first matrix, a second matrix and a noise coefficient matrix; Performing singular value decomposition on the input matrix to obtain a first decomposition matrix, a second decomposition matrix and a third decomposition matrix; Obtain an aliasing coefficient matrix according to the first matrix and the noise coefficient matrix, and obtain a signal source matrix according to the second matrix; According to the first decomposition matrix and the preset matrix, the aliasing coefficient matrix is ​​obtained, and according to the inverse matrix of the preset matrix, the second decomposition matrix and the third decomposition matrix, the signal source matrix is ​​obtained; wherein the preset matrix is ​​determined by a first parameter, a second parameter, a third parameter and a fourth parameter; Determine the first parameter and the second parameter according to the first decomposition matrix, the preset matrix, the first matrix and the noise coefficient matrix; Determine, according to the preset matrix and the preset value set of the fourth parameter, a value set of the third parameter corresponding to the preset value set; Calculate a matrix set of the preset matrix according to the first parameter, the second parameter, the preset value set and the value set; Determine a plurality of groups of signals according to the matrix set, the second decomposition matrix and the third decomposition matrix, each group of signals comprising: a polarization speckle image containing information of the imaging target, and a polarization speckle image containing background information; From the plurality of groups of signals, determine a group of signals having the smallest correlation between the polarization speckle image containing information of the imaging target and the polarization speckle image containing background information as the candidate signals; taking the polarization speckle image containing the information of the imaging target in the selected signal as the obtained polarization speckle image containing the information of the imaging target; Performing autocorrelation processing on the polarization speckle image containing the information of the imaging target to obtain autocorrelation information of the information of the imaging target; According to the autocorrelation information, the Fourier amplitude of the imaging target is obtained by using the Wiener-Schintz theorem; An imaging image of the imaging target is obtained by phase recovery and reconstruction according to the Fourier amplitude.

2. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1, characterized in that: The multiple speckle images include: a speckle image with a polarization azimuth angle of 0°, a speckle image with a polarization azimuth angle of 45°, a speckle image with a polarization azimuth angle of 90°, and a speckle image with a polarization azimuth angle of 135°; the expressions of the first speckle image and the second speckle image are respectively as follows: I=I0+I 90 ; Q=I0-I 90 ; U1=I 45 -I 135 ; Among them, I max represents the first speckle image, I min represents the second speckle image, I represents the first Stokes parameter, Q represents the second Stokes parameter, U1 represents the third Stokes parameter, I0 represents the speckle image with a polarization azimuth angle of 0°, I 45 represents the speckle image with a polarization azimuth angle of 45°, 90 represents the speckle image with a polarization azimuth angle of 90°, I 135 It represents the speckle image with the polarization azimuth angle of 135°.

3. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1, characterized in that: The step of obtaining a speckle polarization difference image according to the first speckle image and the second speckle image comprises: Calculate pixel differences between corresponding pixels of the first speckle image and the second speckle image to obtain the speckle polarization difference image.

4. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1, characterized in that: The relationship between the input matrix, the first matrix, the second matrix, the noise coefficient matrix, the aliasing coefficient matrix, the signal source matrix, the preset matrix, the first decomposition matrix, the second decomposition matrix and the third decomposition matrix is ​​as follows: X=U×D×V T =U×W×W -1 ×D×V T =A×S; A = U × W; S=W -1 ×D×V T ; Wherein, PDI represents the speckle polarization difference image, I min represents the second speckle image, X and All represent the input matrix, represents the first matrix, represents the noise coefficient matrix, represents the first background speckle noise coefficient, represents the second background speckle noise coefficient, represents the second matrix, where I O represents the polarization speckle image containing the information of the imaging target, B represents the polarization speckle image containing background information, A represents the aliasing coefficient matrix, S represents the signal source matrix, U represents the first decomposition matrix, D represents the second decomposition matrix, V T represents the third decomposition matrix, W represents the preset matrix, W -1 represents the inverse matrix of the preset matrix.

5. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1, characterized in that: The process of obtaining the Fourier amplitude of the imaging target according to the autocorrelation information and using the Venatin-Chin theorem is as follows: F -1 {F{O(x,y)} 2 }=O(x,y)☆O(x,y)≈I O (x,y)☆I O (x,y); Wherein, |F{O(x,y)}| represents the Fourier amplitude of the imaging target, O(x,y) represents the information of the imaging target, I O (x, y) represents the obtained polarization speckle image containing the information of the imaging target, and ☆ represents autocorrelation.

6. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1, characterized in that: The spectrum width of the white light broadband light source is 200 nanometers to 400 nanometers.

7. The method for dynamic scattering imaging using the polarization characteristics of speckle field according to claim 1 or 6, characterized in that: The dynamic scattering medium is at least any one of the following: Onion skin, chicken breast tissue, egg shell membrane, turbid liquids, clouds and smoke particles.