Cloud droplet particle phase state identification method and system based on polarization chromatography technology
Through polarization chromatography technology combined with Fresnel equation, the problem of phase state recognition of cloud particles less than 20 μm in the existing technology is solved, and high-precision recognition of cloud microphysical processes is achieved.
Patent Information
- Application Number
- CN202510711781.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-29
- Publication Date
- 2025-08-15
AI Technical Summary
The prior art is difficult to accurately identify the phase state of cloud particles with particle sizes less than 20 μm, which limits the understanding and simulation accuracy of cloud microphysical processes.
Using a method based on polarization tomography technology, the polarization hologram of cloud particles is obtained and digital holographic reproduction is performed by combining polarization imaging with digital holographic imaging. The focus plane is determined using local brightness gradient variance, and the refractive index of particles is calculated in combination with the Fresnel equation to identify their phase state.
It breaks through the resolution limit of existing instruments and can accurately identify the phase state of cloud particles less than 20 μm, improving the recognition accuracy of cloud microphysical processes.
Smart Images
Figure CN120489974A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of polarization detection technology, and in particular to a cloud droplet particle phase state identification method and system based on polarization tomography technology. Background Art
[0002] Cloud precipitation is a key component of the water cycle, significantly impacting weather, climate, weather modification, and atmospheric chemistry. It is also one of the most uncertain factors in weather and climate models. At the microscopic scale, clouds are visible aggregates composed of liquid droplets, ice crystals, or a mixture thereof. Cloud formation, development, and precipitation efficiency are directly governed by the phase state of cloud particles and their interactions. Accurately distinguishing the effects of liquid droplets and ice crystals is crucial for improving climate models and microphysical parameterization schemes, as cloud concentration, size distribution, and interactions not only directly influence cloud optical properties, lifetime, and precipitation efficiency, but also profoundly impact radiation balance and climate change.
[0003] Currently, research methods for identifying cloud particle phases are diverse. Traditional detection methods, such as satellite remote sensing, radar, and aircraft sampling, suffer from issues such as insufficient resolution, difficulty distinguishing particles, and sampling interference. Numerical simulation methods, while capable of revealing in-depth insights into phase transition mechanisms, are limited by model assumptions and computational resources. Thresholding methods are simple to use but rely on subjective parameter settings. Machine learning offers significant advantages in identifying complex cloud states, but it suffers from data dependence, low interpretability, and high computing power requirements. Each of these methods has limitations in detecting small particles in mixed phases. Previous research has demonstrated that digital holography can successfully identify ice crystal shapes. By combining multiple techniques, this approach has addressed the identification of mixed phases of liquid droplets and ice crystals (20 μm-100 μm) in clouds, achieving significant progress and breaking through existing observational bottlenecks, providing important technical support for cloud physics research. However, due to the limited resolution of detection instruments, phase identification of particles smaller than 20 μm remains a bottleneck (it is difficult to distinguish different phases of particles smaller than 20 μm based on particle morphology). The inability to accurately identify the phase state of cloud particles below 20 μm will significantly limit the understanding and simulation accuracy of cloud microphysical processes. Summary of the Invention
[0004] To address the problem of accurate phase identification of particles smaller than 20 μm in the prior art, the present invention provides a cloud droplet particle phase identification method and system based on polarization tomography technology to achieve effective identification of ice crystals and liquid droplets.
[0005] The present invention is achieved through the following technical solutions: A cloud droplet particle phase state identification method based on polarization tomography technology includes the following steps: Step 1: Obtain a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging methods, and obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram in combination with digital holographic reconstruction technology; Step 2: Determine the focus plane of the cloud particles based on the local brightness gradient variance of the cloud particle region in all cloud particle polarization tomograms and the maximum value of the local brightness gradient variance. Determine the polarization information of the cloud particles based on the cloud particle region of interest in the focus plane, the pixel values of the cloud particle region of interest, and the polarization parameters of the light beam. Step 3: Determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the polarization hologram of the cloud particles. Determine the reflection coefficient of the incident light beam based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. Determine the refractive index of the particles based on the reflection coefficient and the Fresnel equation. Determine the phase state of the cloud particles based on the refractive index.
[0006] Preferably, the step 1 of obtaining a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging method includes: The polarization tomography system based on polarization imaging technology and digital holographic imaging principles obtains polarization holographic images of cloud particles and background images without cloud particles. The background image is removed from the polarization holographic image to obtain the polarization hologram of cloud particles.
[0007] Preferably, the polarization tomography imaging system based on polarization imaging technology and digital holographic imaging principle acquires polarization tomography images of cloud particles, including: The incident light beam is converted into linearly polarized light. The linearly polarized light forms refracted light and a main light beam after passing through a beam splitter prism. The polarization information of the refracted light is obtained using a first polarization camera. The main light beam forms a polarization holographic image on a second polarization camera after passing through the particle.
[0008] Preferably, step 1 of obtaining a cloud particle polarization tomogram based on the cloud particle polarization hologram in combination with digital holographic reconstruction technology includes: The cloud particle polarization hologram is subjected to fast Fourier transform and spectrum centering to obtain a centralized spectrum complex matrix, which represents the frequency domain of the polarization hologram. The experimental parameters of the digital holographic reconstruction technology are assigned to the transfer function. Multi-depth reconstruction is performed based on the spectrum complex matrix and transfer function in combination with the angular spectrum method to obtain cloud particle polarization tomograms at different reconstruction distances.
[0009] Preferably, the polarization information of the cloud particles in step 2 includes: The local pixel area of the cloud particle polarization tomogram is determined according to the scale of the cloud particles. The local brightness gradient variance of the local pixel area of each cloud particle polarization tomogram is calculated, and the local pixel area corresponding to the maximum value of the local brightness gradient variance value is used as the focusing plane. The focus plane is normalized, and the particle region of interest is determined based on the brightness of the focus plane. The pixels of the particle region of interest are multiplied with the pixels of the original image using element-by-element multiplication to obtain the pixels of the particle region. The polarization information of the particle is determined based on the pixels of the particle region and the polarization parameters of the light beam.
[0010] Preferably, in step 3, determining the reflection coefficient of the incident light beam according to the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light includes: The scattered light Stokes vector Converted into the electric field component of scattered light; According to the electric field component of the scattered light and combined with the definition of Stokes parameters, the reflection equation is obtained; Write the reflection equation expression as a reflection matrix M , according to the Stokes vector of the scattered light With the reflection matrix M and the Stokes vector of the incident light The relationship between the reflection matrix M Simplify to form a simplified reflection matrix M , according to the reflection matrix M Get the vertical light reflection coefficient of the incident light beam and parallel light reflection coefficient .
[0011] Preferably, the reflection matrix M as follows:
[0012] in, and are the Fresnel reflection coefficients, corresponding to the parallel and perpendicular polarization components respectively; and represent the real and imaginary parts of a complex number respectively; yes The complex conjugate of .
[0013] Preferably, the vertical light reflection coefficient of the incident light beam is and parallel light reflection coefficient as follows:
[0014] in, is the scattered light Stokes vector; is the incident light Stokes vector; Preferably, determining the refractive index of the particles based on the reflection coefficient in combination with the Fresnel equation, and determining the phase state of the cloud particles based on the refractive index includes: The reflection angle is determined by combining the incident angle of the incident light beam with the law of refraction, and the vertical light reflection coefficient is determined by and parallel light reflection coefficient The refractive index of cloud particles is determined by the reflection angle, and the phase state of cloud particles is determined according to the refractive index.
[0015] A cloud droplet particle phase state identification system based on polarization tomography technology, comprising: A polarization tomogram reconstruction module is used to obtain a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging methods, and to obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram and in combination with digital holographic reconstruction technology; A polarization analysis module is used to determine the cloud particle focus plane based on the local brightness gradient variance of the cloud particle region in all cloud particle polarization tomograms and the maximum value of the local brightness gradient variance. The polarization information of the cloud particles is determined based on the cloud particle region of interest in the focus plane, the pixel values of the cloud particle region of interest, and the polarization parameters of the light beam. The phase analysis module is used to determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the cloud particle polarization hologram. The reflection coefficient of the incident light beam is determined based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. The refractive index of the particles is determined based on the reflection coefficient and the Fresnel equation, and the phase state of the cloud particles is determined based on the refractive index.
[0016] Compared with the prior art, the present invention has the following beneficial technical effects: This application provides a method for identifying the phase state of cloud droplet particles based on polarization tomography. This method combines the advantages of polarization imaging and digital holography. By utilizing polarization modulation and detection, it can not only record the three-dimensional amplitude and phase distribution of an object, but also obtain the polarization characteristics of the object. This method enables the characterization of microstructures such as material anisotropy and birefringence, significantly expanding the information acquisition dimension of traditional holography. Traditional observation techniques suffer from significant resolution deficiencies when detecting small particles in mixed phases. This method overcomes the resolution limitations of existing instruments and can accurately identify the phase state of particles (ice crystals and droplets) smaller than 20 μm.
[0017] This application also proposes a cloud droplet particle phase state identification system based on polarization tomography technology, an electronic device and a computer storage medium, which have all the advantages of the above-mentioned cloud droplet particle phase state identification method based on polarization tomography technology. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.
[0019] Figure 1 Schematic diagram of the structure of the polarization tomography system of the present invention; Figure 2 Figure 1 is a flow chart of the polarization tomography technology of the present invention; (a) is the polarization hologram of cloud particles of the present invention, (b) is the background image of the present invention, (c) is the polarization hologram of cloud particles after background removal of the present invention, (d) is the polarization tomography image of cloud particles of the present invention, (e) is the local brightness gradient variance curve of the present invention, (f) is the single particle image located at the focal plane after reconstruction, and (g) is the polarization degree image of a single particle.
[0020] Figure 3 This is the principle diagram of the Fresnel equation of the present invention.
[0021] In the figure, 1. Pulsed laser; 2. Beam expander lens; 3. Collimating lens; 4. Linear polarizer; 5. Beam splitter prism; 6. First polarization camera; 7. Particle to be measured; 8. Microscope objective lens; 9. Second polarization camera. DETAILED DESCRIPTION
[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Generally, the components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations.
[0023] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for protection, but merely represents selected embodiments of the present application. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments in the present application without creative work are within the scope of protection of the present application.
[0024] A cloud droplet particle phase state identification method based on polarization tomography technology includes the following steps: Step 1: Use polarization imaging technology and digital holographic imaging methods to obtain a polarization holographic image of the light beam irradiating the cloud particles and the polarization parameters of the light beam. Remove the background image of the polarization holographic image to obtain a cloud particle hologram. Based on the cloud particle polarization hologram and combined with digital holographic reconstruction technology, obtain a cloud particle polarization tomogram. The specific process is as follows: S1. A polarization tomography system based on the principle of digital holographic imaging is used to obtain the polarization holographic image of cloud particles, the polarization state of the light beam, and the background image without particles.
[0025] The imaging principle of digital holography involves converting a pulsed laser into a parallel, linearly polarized beam through a lens system consisting of a beam expander and a collimator in a coaxial optical path. This parallel light is then modulated into diffracted light after irradiating tiny particles. The parallel light that does not irradiate the cloud particles serves as the reference light, while the diffracted light emitted by the cloud particles serves as the object light. The interference between the object light and the reference light forms circular interference fringes that record information about the cloud particles. These fringes represent a digital hologram of the cloud particles.
[0026] 1) Collect polarization holographic images and polarization parameters of the light beam.
[0027] See Figure 2 Based on the principle of digital holographic imaging, this application designs a polarization tomography imaging system based on the principle of digital holographic imaging, including a pulsed laser 1, a lens group, a linear polarizer 4, a beam splitter prism 5, a first polarization camera 6, a microscope objective 8 and a second polarization camera.
[0028] The pulse laser 1 emits a pulsed laser, which passes through a lens group to form a parallel beam. The parallel beam is modulated into high-purity linear polarized light by a linear polarizer 4 and then irradiated onto a beam splitter prism. The beam splitter prism 5 splits the beam into two paths. The first light beam (refracted light) is irradiated on the first polarization camera 6 , and the first polarization camera obtains the polarization parameters (polarization angle) of the pulsed light beam.
[0029] The first beam (main beam) is incident on the particle to be measured 7. When the beam interacts with the cloud particle, part of the main beam is diffracted by the particle, forming an object beam that carries the cloud particle information. Meanwhile, the main beam, which is not modulated by the cloud particles, serves as the reference beam. The object beam and the reference beam interfere at the microscope objective 8, forming a polarization tomography image. A second polarization camera 9 simultaneously records the polarization holographic images and the Stokes vector at four polarization angles (0°, 45°, 90°, and 135°).
[0030] The polarization camera's sensor is integrated with a micro-polarization array, which includes four polarization directions: 0°, 45°, 90°, and 135°, and can capture polarization images at four polarization angles.
[0031] In some embodiments, the lens group includes a beam expander lens 2 and a collimator lens 3 .
[0032] The pulsed laser is expanded by the beam expander lens 2 to ensure that the beam is evenly distributed in the subsequent optical elements. The collimating lens 3 further collimates the expanded beam to make it a parallel beam, which is incident on the beam splitter prism.
[0033] The pulsed light beam emitted by this polarization tomography system passes through a lens system and then a prism, where it is split into a refracted beam and a main beam. A first polarization camera 6 is used to capture the polarization information of the refracted beam. The main beam illuminates the cloud particles, forming an object beam that carries sample information. The object beam contains key information such as the cloud particle geometry, material properties, and refractive index. Meanwhile, the main beam, unmodulated by the particles, serves as a reference beam, preserving its original polarization state and phase information. The object and reference beams interfere at the microscope objective 8, forming a polarization tomography image on a second polarization camera 9. This polarization tomography system improves lateral resolution to the micrometer level (in the xy plane) and compresses the axial depth of focus to enhance z-axis tomography accuracy. The polarization camera simultaneously records the intensity of the interference field and the Stokes polarization parameter, enabling simultaneous measurement of the particle's three-dimensional morphology and polarization information.
[0034] 2) A polarization tomography system based on the principle of digital holographic imaging is used to obtain a background image without cloud particles. The background image is formed when the pulsed laser does not pass through the cloud particles during the imaging process and is directly captured by the polarization camera.
[0035] During the acquisition process, after the polarization tomography system collects the polarization holographic image, it removes the cloud particles to be measured, and continuously collects multiple images without cloud particles as background patterns. The background images are averaged to obtain a uniform background image.
[0036] For example, five background images are numerically processed and converted into numerical images. Each background image is then added pixel by pixel and averaged to obtain the average value of each pixel position, forming a final average background image to reduce the error caused by background factors.
[0037] S2, removing the average background image of each polarization hologram image to obtain a cloud particle polarization hologram; like Figure 2 As shown in (c), the average background image without cloud particles in the polarization holographic image is removed to obtain a polarization hologram containing only cloud particles. The polarization hologram of cloud particles after removing the background is a two-dimensional matrix in the spatial domain, which represents the distribution of the image in space.
[0038] S3. Obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram and in combination with digital holographic reconstruction technology.
[0039] The numerical reconstruction process is to simulate the real reproduced light waves through numerical methods in the computer, and then extract the basic information of the particles in the polarization hologram.
[0040] The cloud particle polarization hologram is subjected to fast Fourier transform (FFT2) and spectrum centering processing to obtain a frequency domain hologram. Based on the frequency domain hologram and combined with the angular spectrum method, multi-depth reconstruction is performed to obtain cloud particle polarization tomograms at different reconstruction distances.
[0041] In this embodiment, first, the noise is reduced by averaging five frames of background images (1024×1224 pixels), and then the polarization hologram image is read and background subtraction is performed to obtain a net hologram matrix.
[0042] Then, the cloud particle polarization hologram was subjected to fast Fourier transform (FFT2) and spectrum centering processing, and the experimental parameters of the digital holographic reconstruction technology were defined. The experimental parameters were assigned to the transfer function and combined with the angular spectrum method to perform multi-depth reconstruction in the range of 0-170 mm with a step size of 4.86 mm. After normalization, 35 reconstructed polarization tomograms (intensity images at different reconstruction distances) were obtained. The obtained cloud particles were identified by brightness focus judgment using the 35 intensity images at different reconstruction distances.
[0043] The transfer function is as follows:
[0044] Where, is an imaginary unit; To reconstruct the distance; is the wavelength; is the normalized spatial coordinate in the horizontal direction; is the normalized spatial coordinate in the vertical direction.
[0045] The normalization method is as follows:
[0046] Where, is the value of each pixel in the intensity image; is the maximum value of the image matrix; The image matrix is minimized and the pixel values of the image are scaled to the range of [0, 1].
[0047] Step 2: Obtain the local brightness gradient variance of the cloud particle area in all polarization tomograms and construct a local brightness gradient variance curve. Determine the cloud particle focal plane corresponding to each polarization tomography image based on the local brightness gradient variance curve, normalize the focal plane, segment the normalized focal plane based on the threshold, determine the cloud particle region of interest, and determine the particle polarization information based on the pixel values of the cloud particle region of interest and the polarization parameters of the light beam.
[0048] S2.1. Determine the local pixel area of the cloud particle based on its x- and y-direction scales. Calculate the local brightness gradient variance of the local pixel area corresponding to each particle in each polarization tomogram in turn. Use the local pixel area corresponding to the maximum value of the local brightness gradient variance as the focusing plane to obtain the three-dimensional coordinates of the particle.
[0049] In this embodiment, the width and height of the local pixel area are twice the size of the particle in the x (horizontal) and y (vertical) directions, respectively. The local brightness gradient variance of the local pixel area is The calculation method is as follows:
[0050] Where: is the gradient in the local area of the particle; is the average gradient in the local area of the particle; is the set of pixels in the local area; is the number of pixels contained in the local area.
[0051] The Sobel operator is used to determine the gradient as follows:
[0052] Sobel operator and Factor is
[0053] Alternatively, a local brightness gradient variance curve can be constructed based on the local brightness gradient variance of each polarization tomogram, such as Figure 2 As shown in (e), by selecting the maximum value of the curve, the focusing plane of the particle can be obtained, that is, the z-axis position where the particle image is clearest, as shown in Figure 2 (f) As shown in the figure.
[0054] It should be noted that the above operation needs to be performed on images with four different polarization angles (0°, 45°, 90°, and 135°) for each polarization angle image.
[0055] S2.2. Normalize the focus plane, segment the normalized focus plane according to the threshold value, determine the cloud particle region of interest, determine the particle polarization information based on the pixel value of the cloud particle region of interest and the polarization parameter of the light beam, and use pseudo-color images to display the polarization degree distribution, such as Figure 2 As shown in Figure g in the figure, it is convenient to intuitively analyze the polarization characteristic distribution of the target area.
[0056] Normalize the focus plane and scale its pixel value range to the [0,1] interval. Set a threshold between the pixel values [0,1]. Set the pixel values in the image above the threshold to 1, indicating the high-brightness area (target area), and the pixel values below the threshold to 0, indicating the low-brightness area (background area). Threshold segmentation extracts the high-brightness area, and uses element-by-element multiplication to multiply the image above the threshold with the original image, thereby retaining the pixel values of the area of interest (only showing particles) and setting the pixel values of the uninterested area to zero, so that subsequent calculations can focus more on the target area.
[0057] The polarization information of the cloud particles is calculated as follows:
[0058]
[0059]
[0060]
[0061] Where, is the Stokes parameter, is the degree of polarization, is the polarization angle, is the polarization difference represents the total intensity of polarized light, represents the horizontal component of polarized light, Represents the polarization component of polarized light in the 45° direction, Represents the polarization component of polarized light in the circular polarization direction. 、 、 、 Respectively The intensity of linearly polarized light, 、 Represent the intensity values of right-handed and left-handed circularly polarized light, Representatives When linearly polarized light is incident, the Stokes vector Parameters. Step 3: Determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the polarization hologram of the cloud particles. Determine the reflection coefficient of the incident light beam based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. Determine the refractive index of the particles based on the reflection coefficient and the Fresnel equation. Determine the phase state of the cloud particles based on the refractive index.
[0062] S3.1. Determine the Stokes vector of the incident beam (pulsed laser) based on the background imageS in .
[0063] like Figure 1 As shown, during the imaging process of the polarization holographic image, the particle to be measured 7 is removed to obtain a background image without particles, and the Stokes vector of the incident light beam can be determined based on the background image. , which is expressed by the light intensity difference;
[0064] The Stokes vector of the incident beam can also be expressed as:
[0065] in, and are the parallel and perpendicular polarization components of light; and represent the real and imaginary parts of a complex number respectively; yes The complex conjugate of and are the amplitudes of the parallel and perpendicular components; and is the phase of the parallel and perpendicular components.
[0066] S3.2. Determine the Stokes vector of the scattered light formed by the incident light beam passing through the particle based on the polarization holographic image .
[0067] S3.3. Based on the polarization information of cloud particles and the Stokes vector of the incident light beam and the scattered light Stokes vector , determine the vertical light reflection coefficient of the incident light beam and parallel light reflection coefficient , as follows: 1) The scattered light Stokes vector The formula for converting the electric field component of scattered light is:
[0068] 2) According to the electric field component of the scattered light and the definition of Stokes parameters, the reflection equation is obtained, and the reflection matrix is obtained according to the reflection equation M And simplify it to get the simplified reflection matrix M .
[0069] The Stokes parameter is defined as follows:
[0070] The reflection equation is expressed as follows:
[0071] Write the reflection equation expression as a reflection matrix M form:
[0072] According to the Stokes vector of scattered light With the reflection matrix M and the Stokes vector of the incident light The relationship between the reflection matrix M Simplify to form a simplified reflection matrix M .
[0073] The Stokes vector of the scattered light With the reflection matrix M and the Stokes vector of the incident light The relationship is as follows:
[0074] The reflection matrix can be obtained M :
[0075] 3) According to the reflection matrix M Get the vertical light reflection coefficient of the incident light beam and parallel light reflection coefficient .
[0076]
[0077] S3.4. Determine the refractive index of different particles based on the reflection coefficient and the Fresnel equation.
[0078] like Figure 3 As shown, an optically uniform and isotropic single-layer dielectric film has two parallel interfaces, usually the upper one has a refractive index of Air (or vacuum). In the middle is a layer of refractive index The lower layer is a homogeneous medium with a refractive index of The substrate is uniformly attached to the substrate. When a beam of light hits interface 1, multiple reflections and refractions are formed on interface 1 and interface 2 (this principle only considers the case of one refraction and reflection), and each scattered light and refracted light, is the incident angle of light, is the angle of refraction in the medium.
[0079] The electric field vector of a light wave can be decomposed into an electric field component parallel to the incident plane (hereinafter referred to as p component or p wave) and an electric field component perpendicular to the incident plane (hereinafter referred to as s component or s wave). and They represent the p component and s component of the incident light wave respectively.
[0080] The angle of reflection is determined based on the incident angle of the incident light beam and the law of refraction:
[0081]
[0082] The Fresnel equations describe the reflection and transmission behavior of an incident light beam on the surfaces of two different particles. For the angles of incidence and refraction of the cloud particles, the reflection coefficient and Corresponding to s-polarized light and p-polarized light, respectively, according to the vertical light reflection coefficient and parallel light reflection coefficient and the reflection angle determine the refractive index of the cloud particles.
[0083]
[0084]
[0085] The phase states of different cloud particles are determined based on the refractive index of cloud particles.
[0086]
[0087] In this step, the reflection coefficient can be calculated by measuring the Stokes vector of the scattered light and the known polarization state of the incident light. and , and then the refractive index of the target medium can be calculated through the expression of the reflection coefficient. By combining the Fresnel formula and the Stokes vector, the refractive index of the medium can be inferred from the known polarization state of the incident light and the polarization state of the scattered light. This process involves calculating the reflection coefficient, constructing the Mueller matrix, and matching the polarization state of the scattered light with theoretical calculations. Ultimately, by quantitatively analyzing the polarization state of the scattered light, the refractive index of the measured medium can be determined.
[0088] Correspondingly, the present application also provides a cloud droplet particle phase state identification system based on polarization tomography technology, including: A polarization tomogram reconstruction module is used to obtain a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging methods, and to obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram and in combination with digital holographic reconstruction technology; A polarization analysis module is used to determine the cloud particle focus plane based on the local brightness gradient variance of the cloud particle region in all cloud particle polarization tomograms and the maximum value of the local brightness gradient variance. The polarization information of the cloud particles is determined based on the cloud particle region of interest in the focus plane, the pixel values of the cloud particle region of interest, and the polarization parameters of the light beam. The phase analysis module is used to determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the cloud particle polarization hologram. The reflection coefficient of the incident light beam is determined based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. The refractive index of the particles is determined based on the reflection coefficient and the Fresnel equation, and the phase state of the cloud particles is determined based on the refractive index.
[0089] It should be noted that in the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of each module is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules can be combined or integrated into another device, or some features can be ignored or not executed. The modules described as separate components may or may not be physically separated. The components displayed as modules may be one physical unit or multiple physical units, that is, they may be located in one place, or they may be distributed in multiple different places. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.
[0090] In addition, the modules in the various embodiments of the present invention may be integrated into a single processing unit, each module may exist physically separately, or two or more modules may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.
[0091] An electronic device provided in an embodiment of the present application includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the cloud droplet particle phase state identification method based on polarization tomography technology as described in any of the above embodiments.
[0092] Another electronic device provided in an embodiment of the present application may further include: an input port connected to the processor for transmitting multimodal data collected by an external acquisition device to the processor; a display unit connected to the processor for displaying the processing results of the processor to the outside world; and a communication module connected to the processor for enabling communication between the electronic device and the outside world. The display unit may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module includes but is not limited to mobile high-definition link technology (HML), universal serial bus (USB), high-definition multimedia interface (HDMI), wireless connection (including wireless fidelity technology (WiFi), Bluetooth communication technology, low-power Bluetooth communication technology, and communication technology based on IEEE802.11s).
[0093] An embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of the cloud droplet particle phase state identification method based on polarization tomography technology as described in any of the above embodiments are implemented.
[0094] For descriptions of the relevant portions of the polarization tomography-based cloud droplet particle phase identification system, electronic device, and computer-readable storage medium provided in the embodiments of this application, please refer to the detailed description of the corresponding portions of the polarization tomography-based cloud droplet particle phase identification method provided in the embodiments of this application, and these descriptions are omitted here. Furthermore, portions of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art are not described in detail to avoid redundant elaboration.
[0095] The above content is only for explaining the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.
Claims
1. A cloud droplet particle phase identification method based on polarization tomography technology, characterized in that: The following steps are involved: Step 1: Obtain a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging methods, and obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram in combination with digital holographic reconstruction technology; Step 2: Determine the focus plane of the cloud particles based on the local brightness gradient variance of the cloud particle region in all cloud particle polarization tomograms and the maximum value of the local brightness gradient variance. Determine the polarization information of the cloud particles based on the cloud particle region of interest in the focus plane, the pixel values of the cloud particle region of interest, and the polarization parameters of the light beam. Step 3: Determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the polarization hologram of the cloud particles. Determine the reflection coefficient of the incident light beam based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. Determine the refractive index of the particles based on the reflection coefficient and the Fresnel equation. Determine the phase state of the cloud particles based on the refractive index.
2. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 1 is characterized in that: The step 1 of obtaining a polarization hologram of cloud particles based on polarization imaging technology and digital holographic imaging method includes: The polarization tomography system based on polarization imaging technology and digital holographic imaging principles obtains polarization holographic images of cloud particles and background images without cloud particles. The background image is removed from the polarization holographic image to obtain the polarization hologram of cloud particles.
3. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 2 is characterized in that: The polarization tomography imaging system based on polarization imaging technology and digital holographic imaging principle obtains polarization tomography images of cloud particles, including: The incident light beam is converted into linearly polarized light. The linearly polarized light forms refracted light and a main light beam after passing through a beam splitter prism. The polarization information of the refracted light is obtained using a first polarization camera. The main light beam forms a polarization holographic image on a second polarization camera after passing through the particle.
4. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 1 is characterized in that: Step 1: Obtaining a cloud particle polarization tomogram based on the cloud particle polarization hologram and combining it with digital holographic reconstruction technology includes: The cloud particle polarization hologram is subjected to fast Fourier transform and spectrum centering to obtain a centralized spectrum complex matrix, which represents the frequency domain of the polarization hologram. The experimental parameters of the digital holographic reconstruction technology are assigned to the transfer function. Multi-depth reconstruction is performed based on the spectrum complex matrix and transfer function in combination with the angular spectrum method to obtain cloud particle polarization tomograms at different reconstruction distances.
5. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 1 is characterized in that: The polarization information of the cloud particles in step 2 includes: The local pixel area of the cloud particle polarization tomogram is determined according to the scale of the cloud particles. The local brightness gradient variance of the local pixel area of each cloud particle polarization tomogram is calculated, and the local pixel area corresponding to the maximum value of the local brightness gradient variance value is used as the focusing plane. The focus plane is normalized, and the particle region of interest is determined based on the brightness of the focus plane. The pixels of the particle region of interest are multiplied with the pixels of the original image using element-by-element multiplication to obtain the pixels of the particle region. The polarization information of the particle is determined based on the pixels of the particle region and the polarization parameters of the light beam.
6. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 1 is characterized in that: In step 3, the reflection coefficient of the incident light beam is determined based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. The scattered light Stokes vector Converted into the electric field component of scattered light; According to the electric field component of the scattered light and combined with the definition of Stokes parameters, the reflection equation is obtained; Write the reflection equation expression as a reflection matrix M , according to the Stokes vector of the scattered light With the reflection matrix M and the Stokes vector of the incident light The relationship between the reflection matrix M Simplify to form a simplified reflection matrix M , according to the reflection matrix M Get the vertical light reflection coefficient of the incident light beam and parallel light reflection coefficient .
7. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 6 is characterized in that: The reflection matrix M as follows: in, and are the Fresnel reflection coefficients, corresponding to the parallel and perpendicular polarization components respectively; and represent the real and imaginary parts of a complex number respectively; yes The complex conjugate of .
8. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 6 is characterized in that: The vertical light reflection coefficient of the incident light beam and parallel light reflection coefficient as follows: in, is the scattered light Stokes vector; is the incident light Stokes vector.
9. The cloud droplet particle phase identification method based on polarization tomography technology according to claim 6 is characterized in that: Determining the refractive index of the particles based on the reflection coefficient in combination with the Fresnel equation, and determining the phase state of the cloud particles based on the refractive index includes: The reflection angle is determined by combining the incident angle of the incident light beam with the law of refraction, and the vertical light reflection coefficient is determined by and parallel light reflection coefficient The refractive index of cloud particles is determined by the reflection angle, and the phase state of cloud particles is determined according to the refractive index.
10. A cloud droplet particle phase identification system based on polarization tomography technology, characterized in that: include: A polarization tomogram reconstruction module is used to obtain a cloud particle polarization hologram based on polarization imaging technology and digital holographic imaging methods, and to obtain a cloud particle polarization tomogram based on the cloud particle polarization hologram and in combination with digital holographic reconstruction technology; A polarization analysis module is used to determine the cloud particle focus plane based on the local brightness gradient variance of the cloud particle region in all cloud particle polarization tomograms and the maximum value of the local brightness gradient variance. The polarization information of the cloud particles is determined based on the cloud particle region of interest in the focus plane, the pixel values of the cloud particle region of interest, and the polarization parameters of the light beam. The phase analysis module is used to determine the Stokes vector of the incident light beam and the Stokes vector of the scattered light formed by the incident light beam passing through the cloud particles based on the background image and the cloud particle polarization hologram. The reflection coefficient of the incident light beam is determined based on the polarization information of the cloud particles, the Stokes vector of the incident light beam, and the Stokes vector of the scattered light. The refractive index of the particles is determined based on the reflection coefficient and the Fresnel equation, and the phase state of the cloud particles is determined based on the refractive index.