A method, system, electronic device and storage medium for self-calibration of deflection angle

Abstract

The present invention provides a method, system, electronic device, and storage medium for self-calibrating an analyzer angle. The method comprises: obtaining a light intensity dataset, the light intensity dataset comprising multiple sets of light intensity data, each set of light intensity data comprising the true light intensity measurement values ​​of each channel obtained when a polarization imaging system images incident polarized light at any polarization angle; iteratively optimizing the installation deviation angles of multiple analyzers in the polarization imaging system using a preset first objective function and a second objective function to obtain an analyzer angle calibration value; wherein the first objective function performs a first optimization of the installation deviation angle based on the light intensity solution value and the true light intensity measurement value of a redundant response channel; and the second objective function performs a second optimization of the installation deviation angle based on fluctuation information of the obtained polarization degree solution value and / or light intensity solution value of the incident polarized light; and performing analyzer angle self-calibration based on the analyzer angle calibration value. This method can achieve self-calibration even without a known calibration environment.

Description

Technical Field

[0001] The present invention relates to the field of optical imaging technology, and in particular to a method, system, electronic equipment and storage medium for self-calibration of an analyzer angle. Background Art

[0002] Polarization imaging systems are a key technology currently under development. They can be implemented in a variety of ways, including time-sharing, aperture-sharing, amplitude-sharing, and focal-plane-sharing. They can detect both the polarization angle and degree of polarization of a scene. They have widespread applications in areas such as sky polarization navigation, product testing, and military detection.

[0003] For polarization imaging systems, taking a multi-aperture polarization compound-eye system as an example, each aperture uses a sub-eye polarizer with a different angle setting to analyze the incident polarized light. The intensity of each analyzed transmitted light is combined and combined with the Mueller matrix of the multi-aperture system to calculate the polarization degree and polarization direction of the incident polarized light. However, a problem with polarization imaging systems is that the analyzing angles of each aperture polarizer are difficult to install with high precision according to the preset values, which can lead to installation deviations. This causes the actual analyzing angle of each aperture polarizer to deviate from the preset value, resulting in errors in the calculation of the polarization degree and polarization direction of the incident polarized light. Therefore, calibrating the actual analyzing angle of each aperture polarizer is crucial. Currently, supervised methods are commonly used to calibrate the actual analyzing angle of each aperture polarizer. This involves performing supervised learning based on the known polarization angle of the incident polarized light, and then calibrating the analyzing angle error. However, this method requires the use of a standard polarization calibration source with a known polarization angle, which requires high calibration environment conditions and is not conducive to the real-time calibration of the analyzing angle of the polarization imaging system. Summary of the Invention

[0004] The present invention provides a method, system, electronic device and storage medium for self-calibration of the analyzer angle, which are used to solve the problem in the prior art that a standard polarization calibration source with a known polarization angle is required, the calibration environment conditions are high, and it is not conducive to the instant calibration of the analyzer angle of a polarization imaging system.

[0005] The present invention provides a method for self-calibration of an analyzer angle, comprising:

[0006] Acquire a light intensity data set, the light intensity data set comprising: multiple sets of light intensity data, each set of light intensity data comprising a true value of light intensity measurement of each channel obtained when the polarization imaging system images incident polarized light at any polarization angle;

[0007] defining any channel of the polarization imaging system as a redundant response channel, and defining the other channels of the polarization imaging system as channels to be tested;

[0008] Using a preset first objective function and a second objective function, the installation deviation angles of multiple analyzers in a polarization imaging system are iteratively optimized to obtain a calibration value of the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on a light intensity solution value and a true light intensity measurement value of a redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the true light intensity measurement value of the corresponding channel to be tested; the second objective function performs a second optimization on the installation deviation angle based on fluctuation information of a polarization degree solution value and / or fluctuation information of a light intensity solution value of the incident polarized light obtained; the iterative optimization is performed using a discrete parameter optimization method;

[0009] Based on the calibration value of the analyzer angle, perform analyzer angle self-calibration.

[0010] Optionally, the steps of obtaining a light intensity dataset include:

[0011] Arbitrarily rotate the incident polarized light source to multiple unknown polarization angles;

[0012] The multiple analyzers of the polarization imaging system are controlled to perform multiple analyzer imaging on incident polarized light of multiple unknown polarization angles, and the true value of the light intensity measurement of each channel corresponding to the incident polarized light of multiple polarization angles is obtained.

[0013] Optionally, the step of obtaining the light intensity solution value of the redundant response channel includes:

[0014] The polarizer corresponding to the redundant response channel is used as a reference polarizer, and the polarizers corresponding to the multiple channels to be tested are used as polarizers to be tested;

[0015] Obtaining true values ​​of light intensity measurements output by the plurality of polarizers to be analyzed during any polarization analysis and imaging process;

[0016] Based on the true values ​​of the light intensity measurements output by the plurality of polarizers to be analyzed, obtaining the Stokes vector of the incident polarized light at the corresponding polarization angle;

[0017] The Stokes vector is analyzed by using the reference polarizer to obtain a light intensity solution value of the redundant response channel.

[0018] Optionally, the step of first optimizing the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel includes:

[0019] Using the preset first loss function as the first optimization objective;

[0020] Performing a first optimization on the installation deviation angle based on the first optimization objective, the light intensity solution value of the redundant response channel, and the light intensity measurement true value;

[0021] The mathematical expression of the first loss function is:

[0022]

[0023] Among them, loss1 represents the first loss function, w represents the number of groups of light intensity data in the light intensity data set, i em Indicates the calculated value of the redundant response channel corresponding to the mth group of light intensity data, i tm Represents the true value of the light intensity measurement of the redundant response channel in the mth group of light intensity data.

[0024] Optionally, the step of performing a second optimization on the installation deviation angle based on the obtained fluctuation information of the polarization degree solution value and / or the light intensity solution value of the incident polarized light includes:

[0025] Obtaining polarization degree solution values ​​and / or light intensity solution values ​​of incident polarized light at multiple polarization angles;

[0026] The installation deviation angle is secondly optimized based on fluctuation information of polarization degree solution values ​​of incident polarized light at multiple polarization angles and / or fluctuation information of light intensity solution values ​​of incident polarized light at multiple polarization angles.

[0027] Optionally, the step of performing a second optimization on the installation deviation angle based on fluctuation information of a calculated value of a degree of polarization of incident polarized light at multiple polarization angles and / or fluctuation information of a calculated value of a light intensity of incident polarized light at multiple polarization angles includes:

[0028] Using a preset second loss function as a second optimization objective, wherein the second loss function is constructed based on variance or discrete series difference values;

[0029] Based on the second optimization target and the fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized, and / or based on the second optimization target and the fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized.

[0030] Optionally, the steps of performing iterative optimization using discrete parameter optimization include:

[0031] The analyzer corresponding to the redundant response channel is used as a reference polarizer, and the analyzers corresponding to the multiple channels to be tested are used as the polarizers to be tested; a point is randomly selected from a preset multi-dimensional space as an initial test point, the coordinates of the initial test point correspond to the installation deviation angles of the multiple polarizers to be tested, and the dimension of the multi-dimensional space is the same as the number of the polarizers to be tested;

[0032] Based on a preset point set acquisition rule and an initial point to be measured, the point to be measured is expanded, and points that can strongly Pareto dominate the initial point to be measured are selected from the expanded subdivided point set to form a set of points to be measured corresponding to the initial point to be measured, wherein the set of points to be measured includes at least one expanded point to be measured;

[0033] Based on the first objective function, the extended test points are sorted once to obtain a first test point sequence; a predetermined proportion of a portion of the first test point sequence is selected to form a screening point sequence; based on the second objective function, the screening point sequence is sorted again to obtain a second test point sequence;

[0034] Acquire a target test point in the second test point sequence, where the target test point is the test point having the minimum output value of the second objective function corresponding to the extended test points;

[0035] Based on the target test point, the steps of test point expansion, strong Pareto dominance screening, primary sorting, proportional screening, secondary sorting, and target test point re-acquisition are iteratively performed until convergence.

[0036] The present invention also provides a self-calibration system for detecting deflection angle, comprising:

[0037] A sampling module is used to obtain a light intensity data set, wherein the light intensity data set includes: multiple sets of light intensity data, each set of light intensity data includes a true value of the light intensity measurement of each channel obtained when the polarization imaging system images incident polarized light at any polarization angle;

[0038] a pre-defined module, configured to define any channel of the polarization imaging system as a redundant response channel, and define the other channels of the polarization imaging system as channels to be tested;

[0039] An optimization module is configured to iteratively optimize the installation deviation angles of multiple analyzers in a polarization imaging system using a preset first objective function and a second objective function to obtain a calibration value of the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on a light intensity solution value and a true light intensity measurement value of a redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the true light intensity measurement value of the corresponding channel to be tested; and the second objective function performs a second optimization on the installation deviation angle based on fluctuation information of a polarization degree solution value and / or fluctuation information of a light intensity solution value of the incident polarized light obtained; and the iterative optimization is performed using a discrete parameter optimization method.

[0040] The self-calibration module is used to perform self-calibration of the analyzer angle based on the analyzer angle calibration value.

[0041] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any one of the above-described analyzer angle self-calibration methods is implemented.

[0042] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for self-calibration of the analyzer angle as described above is implemented.

[0043] Beneficial effects of the present invention: The present invention provides a method, system, electronic device and storage medium for self-calibration of the analyzer angle, which obtains a light intensity data set, wherein the light intensity data set includes: multiple sets of light intensity data, each set of light intensity data includes the true value of the light intensity measurement of each channel obtained when the polarization imaging system images the incident polarized light of any polarization angle; defines any channel of the polarization imaging system as a redundant response channel, and defines the other channels of the polarization imaging system as channels to be tested; uses the preset first objective function and the second objective function to calibrate the installation deviation angles of multiple analyzers in the polarization imaging system. Iterative optimization is performed to obtain a calibration value for the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on the light intensity solution value and the true light intensity measurement value of the redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the true light intensity measurement value of the corresponding channel to be tested; the second objective function performs a second optimization on the installation deviation angle based on the fluctuation information of the polarization degree solution value and / or the fluctuation information of the light intensity solution value of the incident polarized light; the iterative optimization is performed using a discrete parameter optimization method; and based on the calibration value for the analyzer angle, self-calibration of the analyzer angle is performed. Self-calibration can be achieved without known calibration environment conditions, with high calibration accuracy, and instant calibration of the analyzer angle of the polarization imaging system can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 1 is a flow chart of the analyzer angle self-calibration method provided by the present invention;

[0046] Figure 2 This is a schematic diagram of the process of obtaining the light intensity solution value of the redundant response channel in the analyzer angle self-calibration method provided by the present invention;

[0047] Figure 3 1 is a schematic diagram of a first optimization process in the analyzer angle self-calibration method provided by the present invention;

[0048] Figure 4 1 is a schematic diagram of the second optimization process in the analyzer angle self-calibration method provided by the present invention;

[0049] Figure 5 1 is a flow chart of iterative optimization using a discrete parameter optimization method in the analyzer angle self-calibration method provided by the present invention;

[0050] Figure 6 It is a schematic diagram of the process of obtaining the calibration value of the analyzer angle in the analyzer angle self-calibration method provided by the present invention;

[0051] Figure 7 1 is a schematic diagram of a process for performing self-calibration of the analyzer angle on four analyzers in the first embodiment of the self-calibration method of the analyzer angle provided by the present invention;

[0052] Figure 8 This is a flow chart of the self-calibration process of the four-way analyzer using a supervised learning method;

[0053] Figure 9 Schematic diagram of the imaging mode of the four-aperture polarization imaging system in the second embodiment of the analyzer angle self-calibration method provided by the present invention;

[0054] Figure 10 1 is a schematic structural diagram of the analyzer angle self-calibration system provided by the present invention;

[0055] Figure 11 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0056] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0057] In the following examples, Figures 1-11 The present invention describes the analyzer angle self-calibration method, system, electronic device and storage medium.

[0058] In order to facilitate understanding of the analyzer angle self-calibration method provided by the present invention, the algorithm background of the method is first explained. For the polarization measurement part of the polarization imaging system, whether it is based on the polarization measurement mode of aperture division, amplitude division, or focal plane division, its essence is to use a multi-channel polarizer (polarizer) with different analyzer angles to analyze the incident polarized light from the same target separately, obtain different light intensity responses, and then substitute the known multi-channel polarization Mueller matrix to solve the Stokes vector of the incident light, and then realize polarization imaging. Therefore, it is necessary to calibrate the analyzer angle of the analyzer to determine the Mueller matrix, thereby realizing polarization measurement and polarization imaging with higher polarization accuracy.

[0059] Please refer to Figure 1 This embodiment provides a method for self-calibration of an analyzer angle, including:

[0060] S101: Acquire a light intensity data set, wherein the light intensity data set includes: multiple sets of light intensity data, each set of light intensity data includes a true value of light intensity measurement of each channel obtained when a polarization imaging system images incident polarized light of any polarization angle.

[0061] Specifically, the steps of obtaining the light intensity dataset include:

[0062] S1011: arbitrarily rotating an incident polarized light source by multiple unknown polarization angles. The incident polarized light source can be an existing light source device or polarized light emitted by different polarization imaging scenarios, such as using a camera with a polarization imaging system to photograph any scene and perform polarization imaging.

[0063] S1012: Controlling the multiple analyzers of the polarization imaging system to perform multiple analyzer imaging operations on incident polarized light of multiple unknown polarization angles, thereby obtaining true intensity measurements for each channel corresponding to the incident polarized light at each of the multiple polarization angles. Each set of intensity data includes the true intensity measurements for each channel obtained from analyzing the incident polarized light at any polarization angle. Obtaining this intensity data set facilitates subsequent analyzer angle self-calibration based on each set of intensity data in the intensity data set.

[0064] S102: defining any channel of the polarization imaging system as a redundant response channel, and defining the other channels of the polarization imaging system as channels to be tested.

[0065] It should be mentioned that if there is no installation deviation in the analyzer angle of each channel, then the light intensity solution value of the redundant response channel obtained based on the light intensity measurement true value of the channel to be tested should be equal to the light intensity measurement true value of the redundant response channel. Correspondingly, if the solution value of the redundant response channel is not equal to the light intensity measurement true value of the redundant response channel, it can be determined that the analyzer of each channel to be tested has an installation deviation. Therefore, by defining any channel of the polarization imaging system as a redundant response channel and defining other channels of the polarization imaging system as channels to be tested, it is possible to facilitate the subsequent optimization of the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel, thereby obtaining a more accurate analyzer angle calibration value.

[0066] S103: Using the preset first objective function and second objective function, iteratively optimize the installation deviation angles of multiple polarizers in the polarization imaging system to obtain the calibration value of the polarizer angle; wherein, the first objective function performs a first optimization on the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the light intensity measurement true value of the corresponding channel to be tested; the second objective function performs a second optimization on the installation deviation angle based on the fluctuation information of the polarization degree solution value and / or the fluctuation information of the light intensity solution value of the obtained incident polarized light; the iterative optimization is performed using a discrete parameter optimization method.

[0067] It should be noted that the number of polarizers in the polarization imaging system in this embodiment is 4. In some embodiments, the number of polarizers in the polarization imaging system can be 4 or more, such as 6, 8, etc. The polarization imaging system is composed of 4 linear polarizers or 4 adjacent pixels in a micro-polarization array. The number of polarizers corresponds to the number of channels, and each channel corresponds to a polarizer. The first objective function uses the redundant information self-optimization method as the internal supervision of the polarization angle, and uses the single-channel redundant optimization to install the deviation angle, that is, defining the channel where any polarizer is located as a redundant response channel, and using the output light intensity of the redundant response channel as the redundant item test light intensity. Based on the redundant item test light intensity, the remaining polarizers are polarized without the need for a known standard polarization calibration source.

[0068] Since the minimum value obtained by the first objective function is not unique, it is necessary to meet the requirements of the second objective function on the basis of meeting the first objective function, so as to obtain a unique calibration value of the analyzer angle and improve the calibration accuracy. The second objective function performs a second optimization on the installation deviation angle based on the fluctuation information of the polarization degree solution value and / or the fluctuation information of the light intensity solution value of the incident polarized light. To a certain extent, it can make the installation deviation angles of multiple analyzers approach the true value, thereby obtaining a calibration value of the installation deviation angle with higher accuracy. And based on the calibration value of the installation deviation angle, the calibration value of the analyzer angle is obtained. By adopting the discrete parameter optimization method for dual-objective iterative optimization, the problem of global minimum search in non-convex multivariate function optimization can be effectively solved, and the minimum value distribution of the two non-convex objective functions (the first objective function and the second objective function) can be grasped as a whole, avoiding falling into the local optimum, and finally obtaining a better optimization result with higher accuracy.

[0069] S104: Based on the analyzer angle calibration value, perform analyzer angle self-calibration. It is understandable that the obtained analyzer angle calibration value is assigned to the polarization imaging system to be calibrated, facilitating polarization imaging by the polarization imaging system. This method can effectively implement self-calibration without a known calibration environment. Highly accurate analyzer angle calibration can be performed without introducing sample errors in the polarization angle of the incident polarized light. Furthermore, real-time calibration is achieved during the use of the polarization imaging system, resulting in a high degree of automation, strong feasibility, and low cost.

[0070] Please refer to Figure 2 In some embodiments, the step of obtaining the light intensity solution value of the redundant response channel includes:

[0071] S201: Using the analyzer corresponding to the redundant response channel as a reference polarizer, and using the analyzers corresponding to the multiple channels to be tested as the polarizers to be tested. It should be noted that the output of the reference polarizer is used as a redundant item for testing light intensity. By using the analyzer corresponding to the redundant response channel as the reference polarizer and the output of the reference polarizer as a redundant item for testing light intensity, it is possible to effectively utilize the cross-correlation between multiple apertures or multiple analyzers in a polarization imaging system, thereby achieving self-calibration of the analyzer angle.

[0072] S202: Obtaining true values ​​of light intensity measurements output by the plurality of polarizers to be analyzed during any polarization analysis and imaging process.

[0073] S203: Based on the true values ​​of the light intensity measurements output by the plurality of polarizers to be tested, obtaining the Stokes vector of the incident polarized light at the corresponding polarization angle.

[0074] S204: Analyzing the Stokes vector using the reference polarizer to obtain a light intensity solution value of the redundant response channel. Obtaining the light intensity solution value facilitates first optimization of the installation deviation angle of each analyzer based on the light intensity solution value.

[0075] Please refer to Figure 3 In some embodiments, the step of first optimizing the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel includes:

[0076] S301: Using a preset first loss function as a first optimization objective.

[0077] S302: Perform a first optimization on the installation deviation angle based on the first optimization target, the light intensity solution value of the redundant response channel, and the light intensity measurement true value.

[0078] The following is an explanation of the steps of performing redundant information self-optimization using the first objective function using a specific embodiment:

[0079] When the polarization imaging system to be calibrated has 4 channels (4 channels correspond to 4 apertures and 4 analyzers), first, define any channel of the polarization imaging system as a redundant response channel, define the remaining three channels of the polarization imaging system as channels to be tested, use the analyzer corresponding to the redundant response channel as the reference polarizer, use the analyzer direction of the reference polarizer as the reference direction, and use the output light intensity of the reference polarizer as the redundant test light intensity. Use the analyzers corresponding to the three channels to be tested as the polarizers to be tested. It can be understood that the single-channel redundant optimization target method utilizes the redundant characteristics of the 4-channel polarization solution, that is, when the analyzer angle (installation angle) θ n When (n=1, 2, 3, 4) is the true value (real value), arbitrarily select one of the four channels (here, channel / aperture 4) as the redundant response channel. The analyzer of this redundant response channel is used as the reference polarizer. The calculated result of the incident polarized light output by the reference polarizer is used as the redundant test intensity, that is, the calculated value of the redundant response channel. Then, for polarized light with the same polarization angle, the true value of the intensity measured on the reference polarizer should be the same as the calculated value of the redundant response channel based on the true value of the intensity measured on the remaining three polarizers to be analyzed (corresponding to apertures 1, 2, and 3). If there is a difference, it is due to a deviation in the installation angle of the polarizer to be analyzed. Then, the installation deviation angle is used as the parameter to be optimized and the first optimization is performed.

[0080] Secondly, the three polarizers to be analyzed are used to analyze the incident polarized light of multiple unknown polarization angles to obtain the true value of the light intensity output by the three polarizers to be analyzed. The mathematical expression of this process is:

[0081]

[0082] Where i represents the true value of the light intensity output by the three polarizers to be analyzed, i1, i2, and i3 represent the true value of the light intensity output by polarizers No. 1, 2, and 3 to be analyzed, respectively, E is the expression (1-e) / (1+e) containing the extinction ratio e, θ1, θ2, and θ3 represent the analyzer angles of polarizers No. 1, 2, and 3 to be analyzed, respectively, p represents the unknown degree of polarization of the incident polarized light, I represents the intensity of the incident polarized light, α represents the unknown polarization angle of the incident polarized light, M(θ1,θ2,θ3) represents the expression of the total Mueller matrix containing the three terms θ1, θ2, and θ3, and S represents the Stokes vector of the incident polarized light.

[0083] Then, based on the polarized light intensity, the Stokes vector of the incident polarized light is solved, and the mathematical expression of the Stokes vector of the incident polarized light is obtained as follows:

[0084]

[0085] Here, Q represents the light component of the incident polarized light in the X-axis direction, and U represents the light component of the incident polarized light in the 45° direction.

[0086] Since S carries the installation angle deviation, the Stokes vector is analyzed using the reference polarizer (the total Mueller matrix M4 of the fourth aperture analyzer, M4 = [1E cos 2θ4E sin 2θ4]) to obtain the light intensity solution value of the redundant response channel. The mathematical expression of the light intensity solution value is:

[0087]

[0088] Among them, i e represents the intensity solution of the redundant response channel for incident polarized light, and θ4 represents the analyzer angle of the fourth aperture analyzer. The calculation rule for the D() function is D(a,b,c) = s(a,b) + s(b,c) + s(c,a), where s(a,b) = sin 2(ab).

[0089] If θ1, θ2, θ3, and θ4 used in the above formula are all true values ​​of the system, then i e It should be equal to the measured true value i4 of the 4th aperture (redundant response channel). However, due to the installation deviation of θ1, θ2, θ3, and θ4, i e There is also a deviation from i4. Therefore, the preset first loss function is used to obtain i e The deviation between i and i4 is used, and the preset first loss function is used as the first optimization target. The sum of the squares of the deviations of all w groups of light intensity data (sampling points) is defined as the first loss function.

[0090] The mathematical expression of the first loss function is:

[0091]

[0092] Among them, loss1 represents the first loss function, w represents the number of groups of light intensity data in the light intensity data set, that is, the number of sampling points, i em Indicates the light intensity solution value of the redundant response channel corresponding to the mth group of light intensity data, i tm represents the true intensity measurement value of the redundant response channel in the mth set of light intensity data. Each set of light intensity data includes the true intensity measurement value of each channel when analyzing polarized light at any polarization angle. Specifically, each set of true intensity measurement values ​​includes the true intensity measurement value of one redundant response channel and the true intensity measurement values ​​of multiple channels to be tested. The true intensity measurement values ​​of the multiple channels to be tested are used to calculate the calculated light intensity value of the redundant response channel.

[0093] Set the analyzer angle of the reference polarizer in channel 4 to the reference direction angle of 0, and use the installation deviation angles o(o1, o2, o3, o4) of each analyzer as the parameter to be optimized. Since the analyzer angle of the reference polarizer in channel 4 is the reference direction angle of 0, o4 can be ignored, so the installation deviation angles are o(o1, o2, o3). That is, θ4 = 0, θ1 = -135° + o1, θ2 = -90° + o2, and θ3 = -45° + o3. As can be understood, the analyzer direction of the reference polarizer is used as the main direction, and the analyzer directions of the other polarizers to be analyzed all use the main direction as the 0° direction. The polarization starting and analyzing angles are all based on the main direction.

[0094] Finally, the steps to obtain the calibration value of the installation deviation angle are converted into R 3 Find the optimization problem of minimizing the loss1 function value. Take the first loss function as the first optimization target min{loss1(o)},o∈R 3 Based on the first optimization goal, the installation deviation angle is first optimized.

[0095] It should be noted that when the polarization imaging system to be calibrated has multiple channels (e.g., 6 or 8 channels), the above embodiment can be adaptively modified to achieve the first optimization. For example, any analyzer can be selected as a reference polarizer, the analyzer direction of the reference polarizer can be used as the reference direction, and the output light intensity of the reference polarizer can be used as the redundant test light intensity. The remaining five or seven analyzers can be used as the polarizers to be analyzed, and the above solution method can be used to perform the first optimization of the installation deviation angle.

[0096] Please refer to Figure 4 In some implementations, the step of performing a second optimization on the installation deviation angle based on the obtained fluctuation information of the polarization degree solution value and / or the light intensity solution value of the incident polarized light includes:

[0097] S401: Obtaining polarization degree solution values ​​and / or light intensity solution values ​​of incident polarized light at multiple polarization angles.

[0098] S402: Based on fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles and / or fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, a second optimization of the installation deviation angle is performed. By optimizing the fluctuation stability of the installation deviation angle of each analyzer, the true value of the installation deviation angle is continuously approached, and a unique installation deviation angle value with high accuracy can be obtained.

[0099] In some embodiments, the step of performing a second optimization on the installation deviation angle based on fluctuation information of a calculated value of a degree of polarization of incident polarized light at multiple polarization angles and / or fluctuation information of a calculated value of a light intensity of incident polarized light at multiple polarization angles includes:

[0100] S4022: Using a preset second loss function as a second optimization objective, where the second loss function is constructed based on variance or discrete series difference values.

[0101] S4023: Based on the second optimization target and the fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized, and / or based on the second optimization target and the fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized.

[0102] In order to facilitate understanding of the principles and steps of the above-mentioned second optimization (fluctuation stability optimization), the principles and steps of the fluctuation stability optimization are explained in detail below.

[0103] First, let's explain the reasoning behind optimizing for fluctuation stability. Since the first objective function used in the first optimization is not strictly convex, it's important to determine whether the first objective function has a unique minimum near its true value. Assume that the incident polarized light's polarization angle α (polarization angle) takes on multiple unknown angles, while the polarized light intensity I and polarization degree p remain stable. Assume that the calculated values ​​of the intensity and polarization degree for the optimized parameters o(o1, o2, o3) are I(α) and p(α).

[0104] For any n-th channel θ n The installation deviation angle of the detector is assumed to be different from the true value of the installation deviation angle o n Another minimum point o ne , that is, the second minimum point of the installation deviation angle obtained after the first optimization (it can be understood that there are multiple installation deviation angles that meet the minimum loss function after the first optimization, among which the true value o n represents the first minimum point of the installation deviation angle obtained after the first optimization). The analyzer of the nth channel is at different polarization angles α, with o ne To optimize the parameters, the calculated light intensity value and the true light intensity measurement value satisfy the following conditions:

[0105] i ne =I(α)[1+Ep(α)cos2(α-θ n -o ne )]

[0106] i n =I[1+Ep cos2(α-θ n -o n )]

[0107] Among them, i ne Indicates the light intensity solution value of the nth detector channel under each α, i n Indicates the true value of the light intensity measurement of the nth polarization channel at each α.

[0108] Different minimum point parameters o have the same response under different α, that is, i ne =i n , let φ(sample deflection angle) = α-θ n -o n , expand the above i ne 、i n The expression of is:

[0109] I(α)-I(α)Ep(α)sin2(o n -o ne )sin2φ+I(α)Ep(α)cos 2(o n

[0110] -o ne )cos2φ=I+IEp(α)cos2φ

[0111] The coefficients of the constant terms and cos2φ terms on the left and right sides of the above formula under the parameters to be optimized o(o1,o2,o3) should be equal, and the solution is:

[0112]

[0113] I(α)=I[1+pEtan2(o n -o ne )sin 2φ]

[0114] That is, when the parameter to be optimized o(o1, o2, o3) takes any value and satisfies the above expressions of p(α) and I(α), different parameters to be optimized o(o1, o2, o3) have the same light intensity output, resulting in the inability to find a unique true value in the case of single-channel redundant optimization. Therefore, it is necessary to use the second objective function for further optimization based on the first objective function.

[0115] If and only if o n =o ne When p(α)=p, the polarization solution value does not fluctuate with the sample polarization angle φ. On the contrary, if the polarization solution value fluctuates with the sample polarization angle φ, it means that the solution value is not the true value. Therefore, by using the preset second loss function as the second optimization target, based on the second optimization target and the fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is secondly optimized, and / or based on the second optimization target and the fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is secondly optimized, thereby achieving the determination of a unique value of the installation deviation value of each polarizer.

[0116] In some embodiments, the second loss function is a preset variance loss function or a discrete sequence difference loss function. The mathematical expression of the second loss function is:

[0117]

[0118]

[0119] Among them, loss2 represents the second loss function, represents the variance of the polarization degree solution, represents the variance of the light intensity solution value, ∑|diff(p)| represents the difference of the polarization degree solution value, ∑|diff(I)| represents the difference of the light intensity solution value, p(α m ) represents the degree of polarization corresponding to different polarization angles of incident polarized light.

[0120] The final optimized parameter o(o1,o2,o3) should satisfy both the first and second objective functions, i.e. min{loss1(o(o1,o2,o3))}, min{loss2(o(o1,o2,o3))}.

[0121] It can be understood that the fluctuation stability optimization is due to the fact that the optimal solution of the first objective function is not unique. By constructing a characterization function of the fluctuation, that is, the second objective function, unsupervised learning is performed. When the spatial fluctuation is minimized, the calibration value of the unique installation deviation angle can be determined.

[0122] Furthermore, since the optimization involved in the method for self-calibration of the deflection angle provided by the present invention belongs to non-Pareto non-convex multi-objective optimization, two objective functions (the first objective function and the second objective function) need to be optimized at the same time, and the common linear weighted method in the prior art cannot be well applied to this method. Therefore, this embodiment proposes to use a discrete parameter optimization method to iteratively optimize the first objective function and the second objective function. Compared with the traditional linear weighted method, the discrete parameter optimization method has the following advantages: 1. Avoid falling into local optimality or single objective function optimality; 2. The final result does not depend on the prior weights; 3. The most important goal that should be met first among the multiple goals can be selected. Please refer to Figure 5 In some embodiments, the steps of performing iterative optimization using discrete parameter optimization include:

[0123] S501: Use the analyzer corresponding to the redundant response channel as a reference polarizer, and use the analyzers corresponding to multiple channels to be tested as polarizers to be tested; select any point from a preset multi-dimensional space as an initial test point, the coordinates of the initial test point correspond to the installation deviation angles of the multiple polarizers to be tested, and the dimension of the multi-dimensional space is the same as the number of the polarizers to be tested.

[0124] S502: Based on the preset point set acquisition rules and the initial test point, the test point is expanded. Points that strongly Pareto dominate the initial test point are selected from the expanded subdivided point set to form a test point set corresponding to the initial test point. The test point set includes at least one expanded test point. The expanded test point includes the initial test point. Specifically, the subdivided point set is obtained by discretely subdividing the space surrounding the initial test point. Points that strongly Pareto dominate the initial test point are selected from the subdivided point set. Specifically, only dominant points are considered, and both loss 1 and loss 2 of the dominant points are smaller than those of the initial test point. The selected point set is used as the test point set corresponding to the initial test point.

[0125] S503: Based on the first objective function, the extended test points are sorted once to obtain a first test point sequence; a portion of the first test point sequence with a preset ratio is selected to form a screening point sequence (i.e., a portion of the extended test points with a smaller first objective loss function in the first test point sequence is selected to form a screening point sequence, and the number of screening points can be set according to the ratio based on actual conditions); based on the second objective function, the screening point sequence is sorted again to obtain a second test point sequence.

[0126] S504: Acquire a target test point in the second test point sequence, where the target test point is the test point with the minimum output value of the second objective function corresponding to the extended test points.

[0127] S505: Based on the target test point, iteratively perform the steps of test point expansion, strong Pareto dominance screening, primary sorting, proportional screening, secondary sorting, and target test point re-acquisition until convergence. That is, repeat the above steps S502-S504 until convergence.

[0128] Specifically, let the space where the parameters to be optimized o(o1,o2,o3) are located be R 3 , the initial coordinates of the parameter initialization of the test point are o(0,0,0). Based on the preset point set acquisition rules and the initial test point, the test point is expanded to form the subdivided point set X corresponding to the initial test point. For example: in the 3D space R 3 In the definition of a cube with side length L, each dimension is divided into E points, then the distance between adjacent points is L / E, and the entire space generates a total of E 3 The subdivided point set X of discrete points. The set of points to be measured includes at least one extended point to be measured.

[0129] Then, the first objective function value and second objective function value of each extended test point are calculated. A test point set X0 from X that strongly Pareto dominates the initial test point is selected, i.e., all points in the test point set X0 satisfy loss1(o(o1,o2,o3))≤loss1(o(0,0,0)) and loss2(o(o1,o2,o3))≤loss2(o(0,0,0)). Based on the first objective function value of each extended test point in X0, the extended test points in X0 are sorted to obtain a first test point sequence X1. A predetermined number of partial point sequences from the first test point sequence X1 are selected to form a screening point sequence X′1. For example, based on the first objective function value of each extended test point in X0, the extended test points in X0 are sorted in ascending order to obtain the first test point sequence X1. The top f percentage of extended test points with the smallest first objective function values ​​are selected to form the screening point sequence X′1.

[0130] Based on the second objective function value of each extended test point in X′1, the extended test points in X′1 are re-sorted to obtain a second test point sequence. The target test point (usually the optimal point) in the second test point sequence is selected. At this time, the target test point satisfies the requirement that the second objective function value is minimized while the first objective function value remains small.

[0131] Finally, with the target test point as the center, a new round of test point expansion with a smaller scope, strong Pareto dominance screening, primary sorting, proportional screening, secondary sorting, and re-acquisition of the target test point are performed until convergence. This process is repeated several times, with the cube side length continuously refined. When the total error continues to decrease until convergence or the required number of significant digits of accuracy for the subdivided domain side length is met, the calibration of the installation deviation angle is complete. This discrete parameter optimization method does not rely on the gradient of the objective function and can comprehensively grasp the distribution of the minimum values ​​of the two non-convex objective functions, avoiding falling into local optimality.

[0132] Please refer to Figure 6 In some embodiments, the step of iteratively optimizing the installation deviation angles of multiple analyzers in a polarization imaging system using a preset first objective function and a second objective function to obtain a calibration value of the analyzer angle includes:

[0133] S601: Based on the first objective function and the second objective function, iteratively optimize the installation deviation angles of the multiple analyzers to obtain calibration values ​​of the installation deviation angles corresponding to the multiple analyzers.

[0134] S602: Sum the calibration value of the installation deviation angle and the preset value of the analyzer angle of the corresponding analyzer to obtain the calibration value of the analyzer angle, thereby achieving self-calibration of the analyzer angle with high accuracy and high degree of automation.

[0135] Example 1:

[0136] Please refer to Figure 7 , the self-calibration method for the self-calibration of the polarization angle provided by the present invention is used to self-calibrate the four-way polarizer. First, the unknown polarized light S is analyzed by the four-way polarizer to obtain the true value of the light intensity measurement output by the four-way polarizer (I1, I2, I3, I4), which is used as a set of light intensity data. Arbitrarily rotate the polarized light source or the four-way polarizer system (polarization imaging system), collect the true value of the light intensity measurement output by the four-way polarizer when the incident polarized light of different polarization angles is analyzed and imaged, so as to obtain multiple sets of light intensity data as light intensity data sets. Take the 4th channel as the redundant response channel, the polarizer corresponding to the redundant response channel as the reference polarizer, the other channels as the channels to be tested, and the polarizer corresponding to the channels to be tested as the polarizer to be tested. The polarized light intensity output by the reference polarizer is used as a redundant item to test the light intensity. Substitute the total Mueller matrix of the remaining three polarizers to be tested (the total Mueller matrix is ​​obtained based on the true value of the light intensity measurement of each polarizer to be tested) into any installation error angle (o1, o2, o3), and perform Stokes vector solution of the unknown polarized light S. Use M4 to analyze the Stokes vector and obtain the solution value i of the redundant response channel e , based on the preset first loss function, the solution value i e Compare with the measured true value i4, and use the preset first loss function as the first optimization target. Based on the first optimization target, the first optimization is performed on the installation deviation angle of each polarizer. Then, based on the preset second loss function and the variance of the polarization degree solution value of the incident polarized light under different α (polarization angle of the incident polarized light), the deviation of the polarization degree solution value of the incident polarized light is calculated. Or the variance of the light intensity solution Fluctuation stability is calculated, and the second loss function is used as the second optimization objective. Finally, a discrete parameter optimization method is used to optimize the dual-objective function. This method effectively achieves self-calibration without a known calibration environment, with high calibration accuracy. It also enables real-time calibration of the analyzer angle of the polarization imaging system at a low cost.

[0137] As a comparison, Figure 8 The figure shows the process flow of using supervised learning to perform self-calibration of the four-way analyzer. Figure 8 As shown, the difference between conventional supervised learning methods and the analyzer angle self-calibration provided by the present invention lies in that in the supervised method, the true polarization angle A of the incident polarized light is known. By substituting it into the total Mueller matrix M, the polarization angle α in the Stokes vector solution can be directly obtained. Based on the preset error function, the true polarization angle value, and the polarization angle α in the Stokes vector solution, the installation error angles (o1, o2, o3, o4) can be directly iteratively adjusted. Obviously, the supervised method requires the true polarization angle of the incident polarized light to be known, and cannot effectively calibrate the analyzer angle of the polarization imaging system in real time.

[0138] Example 2:

[0139] Please refer to Figure 9 , Figure 9 A schematic diagram of the imaging mode of a 4-aperture polarization imaging system is shown. Figure 9 The left side shows a stereoscopic image of a 4-aperture polarization imaging system, and the right side shows a cross-sectional image of a front analyzer. As can be seen from the stereoscopic image on the left, the overlapping fields of view of the 4 apertures can perform multi-path simultaneous polarization analysis of the same target. The cross-sectional image of the front analyzer on the right shows the number and polarization direction of each analyzer. Polarizer No. 1 is the main direction, the direction of polarizer No. 2 differs from the main direction by 45°, the direction of polarizer No. 3 differs from the main direction by 90°, and the direction of polarizer No. 4 differs from the main direction by 135°. The polarization angle self-calibration method provided by the present invention is applicable to Figure 9 The analyzer angle self-calibration of the polarization imaging system shown.

[0140] The analyzer angle self-calibration system provided by the present invention is described below. The analyzer angle self-calibration system described below and the analyzer angle self-calibration method described above can be referred to each other.

[0141] Please refer to Figure 10 The analyzer angle self-calibration system provided in this embodiment includes:

[0142] Sampling module 1001 is used to obtain a light intensity data set, wherein the light intensity data set includes: multiple sets of light intensity data, each set of light intensity data includes the true value of the light intensity measurement of each channel obtained when the polarization imaging system images the incident polarized light at any polarization angle;

[0143] A pre-definition module 1002 is configured to define any channel of the polarization imaging system as a redundant response channel, and define other channels of the polarization imaging system as channels to be tested;

[0144] An optimization module 1003 is configured to iteratively optimize the installation deviation angles of multiple analyzers in a polarization imaging system using a preset first objective function and a preset second objective function to obtain a calibration value of the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on a calculated value and a measured true value of a redundant response channel of the incident polarized light, and the second objective function performs a second optimization on the installation deviation angle based on fluctuation information of a calculated value of the degree of polarization and / or a calculated value of the light intensity of the incident polarized light; and the iterative optimization is performed using a discrete parameter optimization method.

[0145] The self-calibration module 1004 is configured to perform self-calibration of the analyzer angle based on the analyzer angle calibration value. The sampling module 1001, pre-definition module 1002, optimization module 1001, and self-calibration module 1002 are connected. This system can achieve self-calibration without a known calibration environment, with high calibration accuracy. It can also achieve real-time calibration of the analyzer angle of the polarization imaging system, with a high degree of automation and strong feasibility.

[0146] In some embodiments, the step of the sampling module 1001 acquiring the light intensity dataset includes:

[0147] Arbitrarily rotate the incident polarized light source to multiple unknown polarization angles;

[0148] The multiple analyzers of the polarization imaging system are controlled to perform multiple analyzer imaging on incident polarized light of multiple unknown polarization angles, and the true value of the light intensity measurement of each channel corresponding to the incident polarized light of multiple polarization angles is obtained.

[0149] In some embodiments, the step of the optimization module 1003 obtaining the light intensity solution value of the redundant response channel includes:

[0150] The polarizer corresponding to the redundant response channel is used as a reference polarizer, and the polarizers corresponding to the multiple channels to be tested are used as polarizers to be tested;

[0151] Obtaining true values ​​of light intensity measurements output by the plurality of polarizers to be analyzed during any polarization analysis and imaging process;

[0152] Based on the true values ​​of the light intensity measurements output by the plurality of polarizers to be analyzed, obtaining the Stokes vector of the incident polarized light at the corresponding polarization angle;

[0153] The Stokes vector is analyzed by using the reference polarizer to obtain a light intensity solution value of the redundant response channel.

[0154] In some embodiments, the optimization module 1003 performs a first optimization on the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel, including:

[0155] Using the preset first loss function as the first optimization objective;

[0156] Performing a first optimization on the installation deviation angle based on the first optimization objective, the light intensity solution value of the redundant response channel, and the light intensity measurement true value;

[0157] The mathematical expression of the first loss function is:

[0158]

[0159] Among them, loss1 represents the first loss function, w represents the number of groups of light intensity data in the light intensity data set, i em Indicates the calculated value of the redundant response channel corresponding to the mth group of light intensity data, i tm Represents the true value of the light intensity measurement of the redundant response channel in the mth group of light intensity data.

[0160] In some embodiments, the step of performing a second optimization on the installation deviation angle based on the obtained fluctuation information of the polarization degree solution value and / or the light intensity solution value of the incident polarized light by the optimization module 1003 includes:

[0161] Obtaining polarization degree solution values ​​and / or light intensity solution values ​​of incident polarized light at multiple polarization angles;

[0162] The installation deviation angle is secondly optimized based on fluctuation information of polarization degree solution values ​​of incident polarized light at multiple polarization angles and / or fluctuation information of light intensity solution values ​​of incident polarized light at multiple polarization angles.

[0163] In some embodiments, the step of performing a second optimization on the installation deviation angle based on fluctuation information of a calculated value of a degree of polarization of incident polarized light at multiple polarization angles and / or fluctuation information of a calculated value of a light intensity of incident polarized light at multiple polarization angles includes:

[0164] Using a preset second loss function as a second optimization objective, wherein the second loss function is constructed based on variance or discrete series difference values;

[0165] Based on the second optimization target and the fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized, and / or based on the second optimization target and the fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized.

[0166] In some embodiments, the steps of performing iterative optimization by the optimization module 1003 using a discrete parameter optimization method include:

[0167] The analyzer corresponding to the redundant response channel is used as a reference polarizer, and the analyzers corresponding to the multiple channels to be tested are used as the polarizers to be tested; a point is randomly selected from a preset multi-dimensional space as an initial test point, the coordinates of the initial test point correspond to the installation deviation angles of the multiple polarizers to be tested, and the dimension of the multi-dimensional space is the same as the number of the polarizers to be tested;

[0168] Based on a preset point set acquisition rule and an initial point to be measured, the point to be measured is expanded, a point that can strongly Pareto dominate the initial point to be measured is selected from the expanded subdivided point set, and a set of points to be measured corresponding to the initial point to be measured is obtained, wherein the set of points to be measured includes at least one expanded point to be measured;

[0169] Based on the first objective function, the extended test points are sorted once to obtain a first test point sequence; a predetermined proportion of a portion of the first test point sequence is selected to form a screening point sequence; based on the second objective function, the screening point sequence is sorted again to obtain a second test point sequence;

[0170] Acquire a target test point in the second test point sequence, where the target test point is the test point having the minimum output value of the second objective function corresponding to the extended test points;

[0171] Based on the target test point, the steps of test point expansion, strong Pareto dominated point screening, primary sorting, proportional screening, secondary sorting, and target test point re-acquisition are iteratively performed until convergence.

[0172] Figure 11 An example of a physical structure diagram of an electronic device is shown below. Figure 11 As shown, the electronic device may include: a processor 1110, a communication interface 1120, a memory 1130, and a communication bus 1140, wherein the processor 1110, the communication interface 1120, and the memory 1130 communicate with each other via the communication bus 1140. The processor 1110 may call the logic instructions in the memory 1130 to execute the analyzer angle self-calibration method.

[0173] In addition, the logic instructions in the above-mentioned memory 1130 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0174] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the self-calibration method of the deflection angle provided by the above methods.

[0175] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to execute the analyzer angle self-calibration method provided by the above methods.

[0176] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0177] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for self-calibration of a deflection angle, characterized in that: include: Acquire a light intensity data set, the light intensity data set comprising: multiple sets of light intensity data, each set of light intensity data comprising a true value of light intensity measurement of each channel obtained when the polarization imaging system images incident polarized light at any polarization angle; defining any channel of the polarization imaging system as a redundant response channel, and defining the other channels of the polarization imaging system as channels to be tested; Using a preset first objective function and a second objective function, the installation deviation angles of multiple analyzers in a polarization imaging system are iteratively optimized to obtain a calibration value of the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on a light intensity solution value and a true light intensity measurement value of a redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the true light intensity measurement value of the corresponding channel to be tested; the second objective function performs a second optimization on the installation deviation angle based on fluctuation information of a polarization degree solution value and / or fluctuation information of a light intensity solution value of the incident polarized light obtained; the iterative optimization is performed using a discrete parameter optimization method; Based on the calibration value of the analyzer angle, perform analyzer angle self-calibration.

2. The analyzer angle self-calibration method according to claim 1, characterized in that: The steps to obtain the light intensity dataset include: Arbitrarily rotate the incident polarized light source to multiple unknown polarization angles; The multiple analyzers of the polarization imaging system are controlled to perform multiple analyzer imaging on incident polarized light of multiple unknown polarization angles, and the true value of the light intensity measurement of each channel corresponding to the incident polarized light of multiple polarization angles is obtained.

3. The analyzer angle self-calibration method according to claim 1, characterized in that: The step of obtaining the light intensity solution value of the redundant response channel includes: The polarizer corresponding to the redundant response channel is used as a reference polarizer, and the polarizers corresponding to the multiple channels to be tested are used as polarizers to be tested; Obtaining true values ​​of light intensity measurements output by the plurality of polarizers to be analyzed during any polarization analysis and imaging process; Based on the true values ​​of the light intensity measurements output by the plurality of polarizers to be analyzed, obtaining the Stokes vector of the incident polarized light at the corresponding polarization angle; The Stokes vector is analyzed by using the reference polarizer to obtain a light intensity solution value of the redundant response channel.

4. The method for self-calibration of the analyzer angle according to claim 1, wherein: The step of first optimizing the installation deviation angle based on the light intensity solution value and the light intensity measurement true value of the redundant response channel includes: Using the preset first loss function as the first optimization objective; Performing a first optimization on the installation deviation angle based on the first optimization objective, the light intensity solution value of the redundant response channel, and the light intensity measurement true value; The mathematical expression of the first loss function is: Among them, loss1 represents the first loss function, w represents the number of groups of light intensity data in the light intensity data set, i em Represents the calculated value of the redundant response channel corresponding to the mth group of light intensity data, i tm Represents the true value of the light intensity measurement of the redundant response channel in the mth group of light intensity data.

5. The analyzer angle self-calibration method according to claim 1, characterized in that: The step of performing a second optimization on the installation deviation angle based on the obtained fluctuation information of the polarization degree solution value and / or the fluctuation information of the light intensity solution value of the incident polarized light includes: Obtaining polarization degree solution values ​​and / or light intensity solution values ​​of incident polarized light at multiple polarization angles; The installation deviation angle is secondly optimized based on fluctuation information of polarization degree solution values ​​of incident polarized light at multiple polarization angles and / or fluctuation information of light intensity solution values ​​of incident polarized light at multiple polarization angles.

6. The analyzer angle self-calibration method according to claim 5, characterized in that: The step of performing a second optimization on the installation deviation angle based on fluctuation information of a polarization degree solution value of the incident polarized light at multiple polarization angles and / or fluctuation information of a light intensity solution value of the incident polarized light at multiple polarization angles includes: Using a preset second loss function as a second optimization objective, wherein the second loss function is constructed based on variance or discrete series difference values; Based on the second optimization target and the fluctuation information of the polarization degree solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized, and / or based on the second optimization target and the fluctuation information of the light intensity solution value of the incident polarized light at multiple polarization angles, the installation deviation angle is second optimized.

7. The analyzer angle self-calibration method according to claim 1, characterized in that: The steps of iterative optimization using discrete parameter optimization include: The analyzer corresponding to the redundant response channel is used as a reference polarizer, and the analyzers corresponding to the multiple channels to be tested are used as the polarizers to be tested; a point is randomly selected from a preset multi-dimensional space as an initial test point, the coordinates of the initial test point correspond to the installation deviation angles of the multiple polarizers to be tested, and the dimension of the multi-dimensional space is the same as the number of the polarizers to be tested; Based on a preset point set acquisition rule and an initial point to be measured, the point to be measured is expanded, and points that can strongly Pareto dominate the initial point to be measured are selected from the expanded subdivided point set to form a set of points to be measured corresponding to the initial point to be measured, wherein the set of points to be measured includes at least one expanded point to be measured; Based on the first objective function, the extended test points are sorted once to obtain a first test point sequence; a predetermined proportion of a portion of the first test point sequence is selected to form a screening point sequence; based on the second objective function, the screening point sequence is sorted again to obtain a second test point sequence; Acquire a target test point in the second test point sequence, where the target test point is the test point having the minimum output value of the second objective function corresponding to the extended test points; Based on the target test point, the steps of test point expansion, strong Pareto dominance screening, primary sorting, proportional screening, secondary sorting, and target test point re-acquisition are iteratively performed until convergence.

8. A self-calibration system for detecting deflection angle, characterized in that: include: A sampling module is used to obtain a light intensity data set, wherein the light intensity data set includes: multiple sets of light intensity data, each set of light intensity data includes a true value of the light intensity measurement of each channel obtained when the polarization imaging system images incident polarized light at any polarization angle; A pre-defined module, configured to define any channel of the polarization imaging system as a redundant response channel, and define other channels of the polarization imaging system as channels to be tested; An optimization module is configured to iteratively optimize the installation deviation angles of multiple analyzers in a polarization imaging system using a preset first objective function and a second objective function to obtain a calibration value of the analyzer angle; wherein the first objective function performs a first optimization on the installation deviation angle based on a light intensity solution value and a true light intensity measurement value of a redundant response channel, and the light intensity solution value of the redundant response channel is obtained based on the true light intensity measurement value of the corresponding channel to be tested; and the second objective function performs a second optimization on the installation deviation angle based on fluctuation information of a polarization degree solution value and / or fluctuation information of a light intensity solution value of the incident polarized light obtained; and the iterative optimization is performed using a discrete parameter optimization method. The self-calibration module is used to perform self-calibration of the analyzer angle based on the analyzer angle calibration value.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the analyzer angle self-calibration method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the analyzer angle self-calibration method according to any one of claims 1 to 7 is implemented.

Images