Polarization camera calibration method

Abstract

The invention discloses a polarization camera calibration method comprising the following steps: constructing a response model of a polarization camera, the response model describing a response value of a pixel of the polarization camera and a relationship between the response value of the pixel and a Stokes vector of incident light, a first modulation matrix of an optical lens and a second modulation matrix of a polarization element of the pixel, incident light is made to enter the polarization camera, the state of the incident light is changed, response values of pixels of the polarization camera are obtained, changing of the state of the incident light comprises changing of the radiation intensity or / and the polarization state or / and the incident angle of the incident light, and a first modulation matrix and a second modulation matrix are obtained through calibration according to measured response value data. According to the method, the response model of the polarization camera is constructed, and the response model considers the modulation effect of the optical lens on the polarization state of the incident light and the modulation effect of the polarization element of the pixel on the polarization state of the incident light, so that the response model better conforms to the actual imaging process of the polarization camera and is more accurate and comprehensive; therefore, the calibration accuracy of the polarization camera is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of optical imaging, in particular to a polarization camera calibration method. BACKGROUND

[0002] Focal plane polarization camera has high imaging accuracy and sensitivity, especially under large field of view conditions, it can effectively capture a wide range of polarization information. The advantage of this camera is that it can record multiple angles and multiple polarization state images at the same time, providing more rich polarization data and providing more information support for accurate polarization parameter calculation. Especially in large-scale polarization imaging applications, focal plane polarization camera can achieve higher measurement efficiency and more accurate polarization imaging results.

[0003] The calibration of the focal plane polarization camera is a key link to ensure its measurement accuracy and imaging quality. However, the existing calibration method for the focal plane polarization camera is not accurate and comprehensive enough. SUMMARY

[0004] The purpose of the present application is to provide a polarization camera calibration method, which improves the accuracy of polarization camera calibration.

[0005] To achieve the above purpose, the present application provides the following technical scheme:

[0006] A polarization camera calibration method, the polarization camera includes an optical lens and a detector, any pixel of the detector is provided with a corresponding polarization element, the incident light is incident to the detector after passing through the optical lens, and the pixel is used to output a response value according to the received light;

[0007] The polarization camera calibration method includes:

[0008] A response model of the polarization camera is constructed, the response model describes the relationship between the response value of the pixel of the polarization camera, the Stokes vector of the incident light, the first modulation matrix of the optical lens and the second modulation matrix of the polarization element of the pixel, the first modulation matrix describes the modulation effect of the optical lens on the radiation intensity and polarization state of the incident light, and the second modulation matrix describes the modulation effect of the polarization element of the pixel on the radiation intensity and polarization state of the incident light;

[0009] The incident light is incident to the polarization camera, and the state of the incident light is changed, and the response value of the pixel of the polarization camera is obtained under each state, wherein changing the state of the incident light includes changing the radiation intensity or / and the polarization state or / and the incident angle of the incident light;

[0010] The first modulation matrix and the second modulation matrix are calibrated according to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response value of the image element of the polarization camera.

[0011] Optionally, the to-be-calibrated parameter of the first modulation matrix comprises a degree of polarization and a polarization angle of the optical lens.

[0012] The first modulation matrix is calibrated according to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response value of the image element of the polarization camera, and comprises:

[0013] The incident light is made to be incident to the polarization camera, and the polarization angle of the incident light is changed, the response value of the image element of the polarization camera is obtained at each polarization angle of the incident light, and the degree of polarization and the polarization angle of the optical lens are obtained according to the multiple polarization angles of the incident light and the response value of the image element corresponding obtained at each polarization angle.

[0014] Optionally, the first modulation matrix is calibrated according to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response value of the image element of the polarization camera, and comprises:

[0015] The incident light is made to be incident to the polarization camera, the incident angle of the incident light is changed, at each incident angle, the polarization angle of the incident light is changed, the response value of the image element of the polarization camera is obtained at each polarization angle of the incident light, and the degree of polarization of the optical lens corresponding to the incident angle is obtained according to the multiple polarization angles of the incident light and the response value of the image element corresponding obtained at each polarization angle.

[0016] The relationship between the degree of polarization of the optical lens and the incident angle of the incident light is obtained according to the multiple incident angles of the incident light and the degree of polarization corresponding to each incident angle.

[0017] Optionally, the to-be-calibrated parameter of the second modulation matrix comprises an extinction ratio of the polarization element of the image element.

[0018] The second modulation matrix is calibrated according to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response value of the image element of the polarization camera, and comprises:

[0019] The incident light is made to be incident to the polarization camera, and the polarization angle of the incident light is changed, the response value of the image element of the polarization camera is obtained at each polarization angle of the incident light, and the extinction ratio of the polarization element of the image element is obtained according to the maximum response value and the minimum response value in the obtained response value of the image element.

[0020] Optionally, the pixels of the detector are arranged in pixel groups, any pixel group comprising at least two pixels corresponding to polarization elements having different polarization angles;

[0021] The modulation effect of the second modulation matrix on the radiation intensity of the incident light is embodied in the relative transmittance of the pixels, which describes the relationship between the response capability of the pixels and the response capability of a reference pixel in the pixel group to which the pixels belong, for the incident light of the same polarization state;

[0022] According to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response values of the pixels of the polarization camera, the second modulation matrix is calibrated to include:

[0023] The incident light is made to be incident on the polarization camera to obtain the response values of the at least two pixels of any pixel group;

[0024] For any pixel of the at least two pixels, a first ratio of the response value of the pixel to the response value of a reference pixel in the at least two pixels is obtained, and the relative transmittance of the pixel is obtained according to the first ratio of the pixel.

[0025] Optionally, the modulation effect of the first modulation matrix on the radiation intensity of the incident light is embodied in the low-frequency relative transmittance of the pixels, which describes the relationship between the transmittance of the optical lens when the incident light is non-perpendicular incident and the transmittance of the optical lens when the incident light is perpendicular incident, for the incident light of the same polarization state;

[0026] According to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response values of the pixels of the polarization camera, the first modulation matrix is calibrated to include:

[0027] The incident light is made to be perpendicular incident on the polarization camera, and the response values of the pixels of a target region of the detector are obtained;

[0028] The incident light is made to be non-perpendicular incident on the polarization camera, and the response values of the pixels of the target region of the detector are obtained;

[0029] A second ratio of the sum of the response values of the pixels of the target region when the incident light is non-perpendicular incident to the sum of the response values of the pixels of the target region when the incident light is perpendicular incident is obtained, and the low-frequency relative transmittance corresponding to the incident angle corresponding to the non-perpendicular incident of the incident light is obtained according to the second ratio.

[0030] Optionally, the target region of the detector is a central field of view region of the detector, which refers to a region on the detector that is within a central field of view of the incident light.

[0031] Optionally, the to-be-calibrated parameter of the response model includes an absolute radiometric coefficient, which describes a relationship between a response capability of the pixel to the incident light and a radiometric intensity of the incident light.

[0032] Calibrating to obtain the absolute radiometric coefficient includes:

[0033] causing the incident light to be incident to the polarization camera, changing a radiometric intensity of the incident light, obtaining a response value of the pixel at each radiometric intensity of the incident light, and obtaining the absolute radiometric coefficient according to the plurality of radiometric intensities of the incident light and the response value of the pixel obtained at each corresponding radiometric intensity.

[0034] Optionally, the device applied includes at least:

[0035] a light source device configured to emit original light;

[0036] a preset polarization element disposed on an emission side of the light source device, configured to cause the original light to pass through the preset polarization element to become polarized light, forming the incident light;

[0037] changing the radiometric intensity of the incident light includes changing a radiometric intensity of the original light emitted by the light source device;

[0038] or / and, changing the polarization state of the incident light includes rotating the preset polarization element around an optical axis of the preset polarization element.

[0039] Optionally, the device applied further includes:

[0040] a first turntable, wherein the polarization camera is disposed on the first turntable, and the first turntable is configured to drive the polarization camera to rotate;

[0041] changing the incident angle of the incident light includes driving the polarization camera to rotate by the first turntable to change the incident angle of the incident light incident to the polarization camera.

[0042] It can be known from the above technical solution that the polarization camera calibration method of the present application constructs a response model of the polarization camera, the response model considers the modulation effect of the optical lens on the polarization state of the incident light and considers the modulation effect of the polarization element of the pixel on the polarization state of the incident light, so that the response model is more consistent with the actual imaging process of the polarization camera, the response model of the polarization camera is more accurate and comprehensive, and therefore the accuracy of the calibration of the polarization camera is improved. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description only aim to some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0044] Figure 1 A flow chart of a polarization camera calibration method provided by an embodiment;

[0045] Figure 2 A schematic diagram of the transmission of incident light to the polarization camera;

[0046] Figure 3 A schematic diagram of a device and a polarization camera arrangement to which the polarization camera calibration method provided by an embodiment is applied.

[0047] The reference signs in the drawings of the specification include:

[0048] 100-polarization camera, 101-light source, 102-integrating sphere, 103-parallel light tube, 104-preset polarization element, 105-second turntable, 106-lens assembly, 107-first turntable. DETAILED DESCRIPTION

[0049] In order to make those skilled in the art better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative effort should fall within the scope of protection of the present application.

[0050] The present embodiment provides a polarization camera calibration method, the polarization camera comprising an optical lens and a detector, any pixel of the detector is provided with a corresponding polarization element, incident light is incident to the detector after passing through the optical lens, and the pixel is used to output a response value according to the received light.

[0051] Reference can be made to Figure 1 , Figure 1 A flow chart of a polarization camera calibration method provided by an embodiment is shown in Figure 1 The polarization camera calibration method comprises the following steps:

[0052] S11: Construct a response model for the polarization camera. The response model describes the relationship between the response value of the pixel of the polarization camera and the Stokes vector of the incident light, the first modulation matrix of the optical lens, and the second modulation matrix of the polarization element of the pixel. The first modulation matrix describes the modulation effect of the optical lens on the radiation intensity and polarization state of the incident light, and the second modulation matrix describes the modulation effect of the polarization element of the pixel on the radiation intensity and polarization state of the incident light.

[0053] Incident light enters the polarization camera, passes through the optical lens, and is imaged onto the detector. Each pixel of the detector is equipped with a corresponding polarization element, and the pixel responds to the light passing through the corresponding polarization element, thereby outputting a response value.

[0054] A response model for a polarization camera is constructed based on polarization optics theory. The imaging process of the polarization camera is quantitatively analyzed, and its imaging process is derived from polarization imaging theory, establishing a complete polarization response model. (For reference...) Figure 2 , Figure 2 This is a schematic diagram illustrating the transmission of incident light to a polarizing camera, as shown below. Figure 2 As shown, the Stokes vector of the incident light is represented as S. in The Stokes vector of the emitted light reaching the detector is represented as S. out Then S out =M P ×M O ×S in M O M represents the first modulation matrix of the optical lens. P The second modulation matrix represents the polarization element of a pixel.

[0055] S12: The incident light is incident on the polarization camera, and the state of the incident light is changed. In each state, the response value of the pixel of the polarization camera is obtained, wherein changing the state of the incident light includes changing the radiation intensity and / or polarization state and / or incident angle of the incident light.

[0056] S13: Based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, the first modulation matrix and the second modulation matrix are calibrated and obtained.

[0057] The first and second modulation matrices are obtained through calibration, thereby determining the response model of the polarization camera.

[0058] The polarization camera calibration method of the embodiment constructs a response model of the polarization camera, the response model considers the modulation effect of the optical lens on the polarization state of the incident light and considers the modulation effect of the polarization element of the pixel on the polarization state of the incident light, so that the response model is more consistent with the actual imaging process of the polarization camera, the response model of the polarization camera is more accurate and comprehensive, and therefore the accuracy of the calibration of the polarization camera is improved.

[0059] In some embodiments, the to-be-calibrated parameters of the first modulation matrix include a degree of polarization and a polarization angle of the optical lens; and the calibration of the first modulation matrix according to the radiation intensity or / and the polarization state or / and the incident angle of the incident light and the response value of the pixel of the polarization camera includes: making the incident light incident to the polarization camera, and changing the polarization angle of the incident light, obtaining the response value of the pixel of the polarization camera at each polarization angle of the incident light, and obtaining the degree of polarization and the polarization angle of the optical lens according to the plurality of polarization angles of the incident light and the response value of the pixel corresponding obtained at each polarization angle.

[0060] The degree of polarization of the optical lens represents the degree of polarization state modulation of the optical lens on the incident light, and is a core polarization performance parameter for quantifying the ability of the optical lens to convert natural light into partially polarized light. The degree of polarization of the optical lens can be represented as the ratio of the radiation intensity of the polarized light in the exit light to the total radiation intensity of the exit light, where the incident light is non-polarized light such as natural light.

[0061] The polarization angle of the optical lens represents the angle between the main polarization direction of the partially polarized light in the exit light and a preset reference direction after the incident light passes through the optical lens, and is a key parameter for representing the polarization direction characteristics of the light after the light is polarized by the optical lens. The main polarization direction of the partially polarized light refers to the direction of vibration of the electric vector of the linear polarization component, and the preset reference direction is, for example, a radial / tangential reference axis in the vertical plane of the optical axis of the optical lens. The polarization angle of the incident light represents the angle between the main polarization direction of the incident light and the preset reference direction when the incident light is polarized light. In some embodiments, the degree of polarization ε1 and the polarization angle θ1 of the optical lens are obtained by fitting the plurality of polarization angles of the incident light and the response value of the pixel corresponding obtained at each polarization angle. For example, least squares fitting can be performed.

[0062] In some embodiments, calibrating the first modulation matrix based on the radiation intensity or / and the polarization state or / and the incident angle of the incident light and the response values of the pixels of the polarization camera comprises: making the incident light incident to the polarization camera, changing the incident angle of the incident light, at each incident angle, changing the polarization angle of the incident light, obtaining the response values of the pixels of the polarization camera at each polarization angle of the incident light, and obtaining the degree of polarization of the optical lens corresponding to the incident angle based on the plurality of polarization angles of the incident light and the corresponding response values of the pixels obtained at each polarization angle; and obtaining the relationship between the degree of polarization of the optical lens and the incident angle of the incident light based on the plurality of incident angles of the incident light and the corresponding degrees of polarization.

[0063] The degree of polarization of the optical lens is related to the incident angle of the incident light, and the degree of polarization of the optical lens can be expressed as a function of the incident angle of the incident light. In the present embodiment, the incident angle of the incident light is changed, and the measured value of the degree of polarization of the optical lens corresponding to each incident angle is obtained, and then the relationship between the degree of polarization of the optical lens and the incident angle of the incident light is obtained based on the plurality of incident angles of the incident light and the corresponding measured values of the degree of polarization.

[0064] In some embodiments, the relationship between the degree of polarization of the optical lens and the incident angle of the incident light can be obtained by fitting based on the plurality of incident angles of the incident light and the corresponding measured values of the degree of polarization. For example, polynomial fitting can be performed, such as fitting with an n-th order polynomial, expressed as:

[0065] ; (1)

[0066] wherein ε1 represents the degree of polarization, θ1 represents the incident angle of the incident light, and Q and E represent coefficients, respectively.

[0067] In some embodiments, the Mueller matrix of the optical lens is expressed as:

[0068] ; (2)

[0069] wherein M represents the Mueller matrix of the optical lens, τ 1 , τ 2 represent the transmittance of the optical lens under the vertical and parallel incident planes in the field of view, respectively, and ε1 represents the degree of polarization of the optical lens.

[0070] The rotation matrix is expressed as:

[0071] ; (3)

[0072] Wherein, A represents a rotation matrix, and θ1 represents a polarizing angle of the optical lens.

[0073] The first modulation matrix of the optical lens is represented as:

[0074] ; (4).

[0075] Wherein, M O represents the first modulation matrix of the optical lens, the degree of polarization and the polarizing angle corresponding to different fields of view in the optical system are different, therefore, the Mueller matrix M of the optical lens is multiplied by a rotation matrix on both sides, that is, A(-θ1)·M·A(θ1), to obtain the first modulation matrix of the optical lens.

[0076] In some embodiments, the to-be-calibrated parameter of the second modulation matrix includes an extinction ratio of a polarization element of a pixel; and the calibration to obtain the second modulation matrix includes: making the incident light incident to the polarization camera, and changing a polarization angle of the incident light, obtaining a response value of the pixel of the polarization camera at each polarization angle of the incident light, and obtaining the extinction ratio of the polarization element of the pixel according to a maximum response value and a minimum response value in the obtained response value of the pixel.

[0077] The extinction ratio of the polarization element describes the ratio between a useful polarization component and an unnecessary polarization component after the incident light passes through the polarization element, and reflects the suppression ability of the polarization element to the unnecessary polarization state, and a high extinction ratio indicates that the polarization element can effectively isolate the unnecessary polarization state. The polarization angle of the incident light represents an included angle between a main polarization direction of the incident light and a preset reference direction when the incident light is polarized light. The main polarization direction of the polarized light refers to a direction of vibration of an electric vector of a linear polarization component, and the preset reference direction is, for example, a radial / tangential reference axis in a vertical plane of an optical axis of the incident light. In some embodiments, the extinction ratio of the polarization element of the pixel is obtained by calculating a ratio between the maximum response value and the minimum response value in the obtained response value of the pixel. In some embodiments, the second modulation matrix is represented as:

[0078] ; (5)

[0079] Wherein, M P represents the second modulation matrix of the polarization element of the pixel, t 2 x represents a maximum transmittance of the polarization element of the pixel, ε2 2 represents the extinction ratio of the polarization element of the pixel, ε2 2 =t 2 y / t 2 x , t2 y the minimum transmittance of the polarization element of the pixel, and θ2 represents the polarization angle of the polarization element of the pixel.

[0080] In some embodiments, the pixels of the detector are arranged in pixel groups, any pixel group comprising at least two pixels, the polarization elements of the at least two pixels corresponding to different polarization angles. In this way, the at least two pixels of a pixel group respectively respond to light of different polarization angles in the incident light. Exemplarily, a pixel group comprises four pixels.

[0081] In some embodiments, the modulation effect of the second modulation matrix on the radiation intensity of the incident light is embodied in the relative transmittance of the pixel, which describes the relationship between the response capability of the pixel and the response capability of a reference pixel in the pixel group to which the pixel belongs, for the incident light of the same polarization state; and the response value of the pixel of the polarization camera is calibrated to obtain the second modulation matrix according to the radiation intensity or / and the polarization state or / and the incident angle of the incident light.

[0082] S21: causing the incident light to be incident on the polarization camera to obtain the response value of the at least two pixels of any pixel group;

[0083] S22: for any pixel of the at least two pixels, obtaining a first ratio of the response value of the pixel to the response value of a reference pixel in the at least two pixels, and obtaining the relative transmittance of the pixel according to the first ratio of the pixel.

[0084] In the polarization camera, the polarization elements corresponding to the pixels of each pixel group are different, which will inevitably affect the radiation intensity of the light passing through the polarization element and the response capability of the corresponding pixel.

[0085] In some embodiments, the incident light can be caused to be normally incident on the polarization camera, i.e., the incident angle of the incident light is 0°, and the response value of the at least two pixels of the pixel group in the central field of view of the detector is obtained accordingly, and then the relative transmittance of the pixel is obtained. The pixel group in the central field of view of the detector refers to the pixel group in the central field of view of the incident light on the detector. Among them, the radiation intensity and the polarization state of the incident light are determined.

[0086] Exemplarily, the relative transmittance of the pixel can be represented as:

[0087] ; (6)

[0088] wherein, T k represents the relative transmittance of the kth pixel in the pixel group, DN k l,prepresents the response value of the kth pixel of the pixel group (l, p), DC k l,p represents the dark current value of the kth pixel of the pixel group (l, p), DN 1 l,p represents the response value of the reference pixel of the pixel group (l, p), DC 1 l,p represents the dark current value of the reference pixel of the pixel group (l, p). The pixel group (l, p) is the pixel group of the detector in the central field of view.

[0089] In some embodiments, the modulation effect of the first modulation matrix on the radiation intensity of the incident light is embodied in the low-frequency relative transmittance of the pixel, which describes the relationship between the transmittance of the optical lens when the incident light is not perpendicular to the incident light and the transmittance of the optical lens when the incident light is perpendicular to the incident light for the same polarization state of the incident light.

[0090] According to the radiation intensity or / and polarization state or / and incident angle of the incident light and the response value of the pixel of the polarization camera, calibrating the first modulation matrix comprises the following steps:

[0091] S31: making the incident light perpendicular to the polarization camera, and obtaining the response value of the pixel of the target region of the detector;

[0092] S32: making the incident light non-perpendicular to the polarization camera, and obtaining the response value of the pixel of the target region of the detector;

[0093] S33: obtaining the second ratio of the sum of the response values of the pixel of the target region when the incident light is non-perpendicular to the sum of the response values of the pixel of the target region when the incident light is perpendicular, and obtaining the low-frequency relative transmittance corresponding to the incident angle corresponding to the non-perpendicular incident light of the incident light according to the second ratio.

[0094] Different incident angles of incident light result in different transmittances of optical lenses, which leads to non-uniform light radiation intensity received by the image plane under different incident angles of incident light. By calibrating the low-frequency relative transmittance of the pixel, the influence of the incident angle of the incident light on the response ability of the pixel is obtained. In this embodiment, the relationship between the response value of the pixel of the target region of the detector when the incident light is non-perpendicular and the response value of the pixel of the target region of the detector when the incident light is perpendicular represents the relationship between the transmittance of the optical lens when the incident light is non-perpendicular and the transmittance of the optical lens when the incident light is perpendicular.

[0095] The target region of the detector can be a central field of view region of the detector, which refers to a region on the detector within a central field of view of incident light. When the incident light is vertically incident to the polarization camera, the central field of view region of the detector is located in a central region of the detector. When the incident light is non-vertically incident to the polarization camera, i.e., the incident angle of the incident light is greater than 0°, the central field of view region of the detector deviates from the center of the detector.

[0096] In some embodiments, the pixels of the detector are arranged in pixel groups, each pixel group including at least two pixels, and obtaining the response value of the pixels in the target region of the detector includes: for each pixel group in the target region of the detector, obtaining a weighted mean of the response values of the at least two pixels in the pixel group as the response value of the pixel group. The second ratio of the sum of the response values of the pixel groups in the target region when the incident light is non-vertically incident to the sum of the response values of the pixel groups in the target region when the incident light is vertically incident is used to obtain the low-frequency relative transmittance corresponding to the incident angle corresponding to the non-vertically incident incident light.

[0097] For example, a pixel group includes four pixels, and the orthogonality of trigonometric functions is used to calculate the light intensity by a weighted mean of the responses of the four detection channels. The relative change of the light intensity is calculated by the following formula using the response values (numerator) of the four detection channels of the m*m pixel region of the measured field of view and the response values (denominator) of the m*m pixel region of the central field of view, i.e., the low-frequency relative transmittance is obtained. The m*m pixel region of the central field of view is the target region of the detector when the incident light is vertically incident, and the m*m pixel region of the measured field of view is the target region of the detector when the incident light is non-vertically incident.

[0098] ; (7)

[0099] wherein P (l, p) represents the spatial low-frequency relative transmittance, X 1 l,p , X 2 l,p , X 3 l,p , X 4 l,p respectively represent the response values of the four pixels of the pixel group (l, p), which are response values after data processing.

[0100] In some embodiments, the to-be-calibrated parameter of the response model includes an absolute radiometric coefficient, which describes a relationship between a response capability of the pixel to the incident light and a radiometric intensity of the incident light. The calibration of the absolute radiometric coefficient includes: making the incident light incident to the polarization camera, changing the radiometric intensity of the incident light, obtaining a response value of the pixel at each radiometric intensity of the incident light, and obtaining the absolute radiometric coefficient according to the plurality of radiometric intensities of the incident light and the response value of the pixel obtained at each corresponding radiometric intensity. The absolute radiometric coefficient can reflect a quantitative relationship between a response value of a pixel output of the polarization camera and the radiometric intensity of the incident light.

[0101] In some embodiments, the response value of the pixel of the target region of the detector is obtained at each radiometric intensity of the incident light, and the target region of the detector is a central field of view region of the detector. Since the degree of polarization of the optical lens is equal to zero when the incident light is normally incident to the polarization camera, the incident light can be normally incident to the polarization camera, the radiometric intensity of the incident light can be changed, and the response value of the pixel can be obtained at each radiometric intensity of the incident light to obtain the absolute radiometric coefficient, so that the influence of the optical lens on the calibration of the absolute radiometric coefficient can be avoided.

[0102] The response value of the light radiation incident in the central field of view to the polarization camera is represented as:

[0103] ; (8)

[0104] wherein A represents the absolute radiometric coefficient, T a represents a relative transmittance of the polarization element of the pixel with the a-th polarization angle, P a (l, p) represents a low-frequency relative transmittance of the optical lens, I represents the radiometric intensity of the incident light, and C represents a dark current coefficient, which can also be referred to as a dark current correction system.

[0105] Since the degree of polarization of the optical lens is equal to zero when the incident light is normally incident, i.e., P a (l, p) = 1, the integration time, the basic gain level, and the like are set, and the polarization camera is controlled to take an n×n pixel window in the central field of view image plane, and the average of the gray scale values corresponding to the pixel window is:

[0106] (9)

[0107] It can be obtained that: .

[0108] In some embodiments, the to-be-calibrated parameter of the response model further includes a dark current coefficient; and calibrating the obtained dark current coefficient includes: blocking the incident light from being incident to the polarization camera, obtaining the response value of the pixel of the polarization camera for multiple times, and obtaining the dark current coefficient of the pixel according to the response values of the pixel obtained for multiple times. The dark current of the detector is mainly caused by the electronic device background signal and the uncertainty noise. In some embodiments, the response values of the pixel obtained for multiple times can be averaged to obtain the dark current coefficient of the pixel.

[0109] According to the polarization radiation transfer theory, the Stokes vector of the incident light is represented as S in =[I Q U] T , wherein I, Q and U are three parameters of the Stokes vector of the incident light. The Stokes vector of the outgoing light reaching the detector is S out =M P ×M O ×S in , and S out is expanded to be:

[0110] (10)

[0111] The radiation intensity of the light received by the detector is represented as:

[0112] (11)

[0113] The response value of the output of the pixel of the detector is represented as:

[0114] ;(12)

[0115] , wherein A represents the absolute radiation coefficient, T a represents the relative transmittance of the polarization element of the pixel with the a-th polarization angle, P a (l, p) represents the low-frequency relative transmittance of the optical lens, I represents the radiation intensity of the incident light, and C represents the dark current coefficient, which can also be referred to as a dark current correction system.

[0116] After the calibration is completed, the calibration result can also be verified. The calibrated polarization camera is compared with the standard test scene with a known polarization state to verify the calibration result. By measuring the known polarized light source, it is ensured that the calibrated camera can accurately measure the degree of polarization and the polarization angle.

[0117] Result optimization: according to the verification result, the parameters in the polarization response model are further adjusted to ensure that the model is highly consistent with the actual data, and the accurate calibration of the polarization camera is finally completed.

[0118] Through data calculation and model optimization, the precise values of the key parameters of the polarization camera, such as the degree of polarization and the polarization angle, are finally obtained, and the polarization calibration is completed. At this time, the calibration accuracy and imaging quality of the split focal plane polarization camera are significantly improved.

[0119] In some embodiments, the device to which the polarization camera calibration method is applied at least includes: a light source device for emitting original light; and a preset polarization element arranged on the light exit side of the light source device for making the original light into polarized light after passing through the preset polarization element to form incident light. The light source device emits original light, and the original light becomes polarized light after passing through the preset polarization element, thereby forming incident light, and the incident light is incident on the polarization camera. The preset polarization element can be, but is not limited to, a wire grid polarizer.

[0120] In some embodiments, the light source device includes: a light source for emitting light and arranged on the light entrance of an integrating sphere; and an integrating sphere provided with a light entrance and a light exit, the light entrance being used for allowing the light emitted by the light source to enter the integrating sphere, and the light exit being used for allowing the light in the integrating sphere to exit. The use of the integrating sphere makes the original light emitted by the light source device uniform in intensity, and can make the formed incident light uniform in intensity. The light source can be, but is not limited to, a halogen lamp.

[0121] In some embodiments, the device to which the polarization camera calibration method is applied further includes: a collimator arranged between the light source device and the preset polarization element for making the original light emitted by the light source device into parallel light to simulate incident light from infinity.

[0122] In some embodiments, the device to which the polarization camera calibration method is applied further includes: a first rotary table on which the polarization camera is arranged and which can drive the polarization camera to rotate. In some embodiments, the device to which the polarization camera calibration method is applied further includes: a second rotary table on which the preset polarization element is arranged and which is used to drive the preset polarization element to rotate. In some embodiments, the device to which the polarization camera calibration method is applied further includes: a lens assembly arranged between the preset polarization element and the polarization camera.

[0123] For example, reference can be made to Figure 3 , Figure 3 The schematic diagram of the device to which the polarization camera calibration method provided by an embodiment is applied and the arrangement of the polarization camera is shown in Figure 3 The light source 101 is arranged on the light entrance of the integrating sphere 102, the light exit of the integrating sphere 102 emits light, the collimator 103 is arranged between the light source device and the preset polarization element 104, and the lens assembly 106 is arranged on the side of the preset polarization element 104 away from the collimator 103. The polarization camera 100 is arranged on the first rotary table 107, and the preset polarization element 104 is arranged on the second rotary table 105.

[0124] In some embodiments, the changing the radiation intensity of the incident light comprises changing the radiation intensity of the original light emitted by the light source device. Changing the radiation intensity of the original light emitted by the light source device can correspondingly change the radiation intensity of the formed incident light.

[0125] In some embodiments, the changing the polarization state of the incident light comprises rotating the preset polarization element 104 around the optical axis of the preset polarization element 104. Rotating the preset polarization element 104 around the optical axis thereof can change the polarization angle of the incident light formed after passing through the preset polarization element 104.

[0126] In some embodiments, the applied device further comprises a first turntable 107, wherein the polarization camera 100 is arranged on the first turntable 107, and the first turntable 107 is used to drive the polarization camera 100 to rotate. The changing the incident angle of the incident light comprises driving the polarization camera 100 to rotate by the first turntable 107 to change the incident angle of the incident light incident to the polarization camera 100. The posture of the polarization camera 100 can be adjusted by the first turntable 107, which can include adjusting the pitch angle, the azimuth angle or the roll angle of the polarization camera 100. The first turntable 107 can be a multi-dimensional first turntable.

[0127] In the experimental environment and device preparation stage, first, a calibration platform is built in a darkroom to ensure that there is no external stray light interference. The polarization camera 100 is preheated for 10 minutes after being turned on, a dark field image is collected, and the average dark current is calculated to ensure that the device is in a stable state.

[0128] In the experiment, the rotation of the preset polarization element 104 and the polarization camera 100 is controlled by a high-precision turntable and a multi-dimensional turntable. First, the posture of the camera, including the pitch angle, the azimuth angle and the roll angle, is adjusted by the multi-dimensional turntable to ensure that the camera can cover different field of view regions. Then, the high-precision turntable is used to rotate the preset polarization element 104 to adjust the angle thereof, which is accurately controlled from 0° to 180°. When rotating each time, the preset polarization element 104 stays for 2 seconds to ensure that the light intensity is stable before triggering the camera to collect images. In this way, the angle change of the preset polarization element 104 and the posture adjustment of the camera can be performed synchronously to ensure that image data of different polarization directions and field angles are obtained.

[0129] The polarization camera of the embodiment can be a split focal plane polarization camera. Accurate calibration of the split focal plane polarization camera is a key link to ensure its measurement accuracy and imaging quality. Traditional calibration methods mainly rely on single-angle measurement and do not fully consider the influence of optical lenses on polarization parameters, limiting the comprehensiveness and accuracy of calibration. Under the condition of large field of view imaging, the polarization effect of the optical lens will cause the measurement error of the polarization degree and the polarization angle to accumulate, affecting the overall imaging quality and data reliability. Therefore, how to effectively compensate for the polarization effect of the optical lens and improve the correction accuracy of the polarization camera in complex environments has become an important problem to be solved. The embodiment proposes a high-precision calibration method and device for a large field of view split focal plane polarization camera, and establishes a new polarization imaging model. The present application introduces a multi-angle polarization measurement method, which measures the polarization state at different incident angles to improve the calculation accuracy of the polarization parameters, making the calibration results more consistent with the actual imaging conditions. This method can effectively improve the calibration accuracy and imaging quality of the polarization camera, and is especially suitable for large field of view imaging systems, providing theoretical support and technical support for high-precision polarization measurement. In addition, the method combines radiation calibration and polarization calibration to improve the reliability of polarization parameter calculation, thereby improving the calibration accuracy of the polarization camera.

[0130] The method aims to improve the polarization calibration accuracy of the split focal plane polarization camera. The method first establishes a mathematical model considering the polarization effect of the optical lens, which describes the influence of the optical system on the polarization parameters. Then, by measuring the polarization state at different incident angles, multi-angle polarization data is collected to support the accurate calculation of the polarization parameters. Based on the traditional polarization imaging and calibration technology, the method introduces the polarization effect of the optical lens, combines multi-angle measurement and optical system modeling, and effectively improves the calibration accuracy of the polarization camera. The specific content is as follows:

[0131] 101: Based on the theory of polarization optics, a calibration model is first established for the optical characteristics of the split focal plane polarization camera. The model quantifies key parameters such as optical lens transmittance, degree of polarization, and polarization principal axis direction by analyzing the polarization modulation process of light in the camera system, combining Mueller matrix and Stokes parameters, and forming a mathematical equation describing the change of polarization state from incident light to detector, providing theoretical support for subsequent experiments.

[0132] 102: To realize multi-angle polarization data acquisition, a calibration platform including a multi-dimensional turntable and a high-precision polarizer control system needs to be built. The multi-dimensional turntable adjusts the pitch angle, azimuth angle, and other spatial attitudes of the camera to simulate large field of view imaging conditions; at the same time, the high-precision turntable rotates the polarizer step by step to generate a sequence of incident light with different polarization directions. By synchronously triggering the camera to capture images, the original gray value (DN value) and corresponding polarizer angle parameters at different incident angles are recorded to construct a multi-dimensional data set covering the full field of view.

[0133] 103: Preprocess the collected raw data to eliminate noise and extract valid signals. Remove abnormal pixel values by 3σ criterion, apply median filtering and Gaussian filtering to smooth image noise, and normalize the data. For unknown parameters in the polarization calibration model (such as lens transmittance, degree of polarization, etc.), select the effective data area (usually the center area of the image) that meets the signal-to-noise ratio requirement to provide high-quality input for subsequent parameter inversion.

[0134] 104: When solving the absolute radiometric calibration coefficient, the radiometric calibration experiment needs to be carried out simultaneously. Calibrate the radiometric response characteristics of the camera by using the integrating sphere uniform light source, and remove the invalid data with weak or overexposed response of the built-in spectrometer. Perform dark background subtraction (eliminate detector dark current noise) and flat field correction (compensate for uneven illumination) on the valid data to ensure the mapping relationship between radiance and DN value is accurate and reliable, and provide a standardized radiometric reference for polarization parameter calculation.

[0135] 105: Based on the preprocessed data, use optimization algorithms such as least squares fitting to iteratively solve the unknown parameters in the polarization calibration model, including lens polarization modulation parameters, absolute radiometric calibration coefficients, and dark current correction coefficients. Through multi-angle data weighted fusion, the final output is the globally optimal Stokes parameters (I, Q, U, V), degree of polarization, and polarization angle distribution. After parameter verification, embed the calibration results into the camera system to realize real-time polarization imaging correction under large field of view conditions, marking the completion of the whole process calibration of the split focal plane polarization camera.

[0136] The present application can achieve the following technical effects:

[0137] 1. Improve the calibration accuracy of large field of view polarization cameras. Existing polarization camera calibration methods usually rely on traditional single-angle polarization measurement, and often ignore the influence of optical lenses on polarization parameters. Under large field of view conditions, this neglect may cause deviation in the measurement results, affecting the calibration accuracy. This method introduces the polarization effect of optical lenses and combines multi-angle measurement method, fully considers the influence of optical system and field of view angle, thereby significantly improving the measurement accuracy of polarization degree and polarization angle. This makes the calibration results more realistic, and can better reflect the polarization characteristics under actual imaging environment.

[0138] 2. Enhanced reliability of polarization measurement. Existing methods often rely on single-angle or limited field-of-view measurements, which may not accurately capture the polarization variations across the entire imaging area. Especially in large field-of-view imaging systems, the loss or error of polarization information can be significant. With the multi-angle polarization measurement method in this patent, the camera field-of-view angle is adjusted using a multi-dimensional turntable, while the polarizer is precisely controlled, allowing comprehensive polarization data collection across the entire field-of-view range. This significantly reduces measurement errors caused by changes in field-of-view angle, improving the reliability of polarization measurement.

[0139] 3. Optimized application performance of polarization camera. Since existing polarization calibration methods often ignore the polarization effect of optical lenses or only focus on single-angle polarization states, the imaging quality of the camera cannot meet the high-precision requirements in actual applications, especially in complex optical systems. This method introduces modeling of the polarization effect of the optical system and multi-angle data fusion, optimizing the overall application performance of the polarization camera. Whether in different lighting conditions or in complex optical systems (such as systems with multiple lenses and polarizers), this method can provide more stable and reliable polarization measurement results, making the polarization camera perform more accurately and efficiently in actual applications.

[0140] The above describes in detail a polarization camera calibration method provided by the present application. In this paper, specific examples are used to explain the principles and implementation of the present application. The above examples are only used to help understand the method and core idea of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. A polarization camera calibration method, characterized in that, The polarization camera includes an optical lens and a detector. Each pixel of the detector is equipped with a corresponding polarization element. Incident light passes through the optical lens and then enters the detector. The pixel is used to output a response value based on the received light. The polarization camera calibration method includes: A response model of the polarization camera is constructed. The response model describes the relationship between the response value of the pixel of the polarization camera and the Stokes vector of the incident light, the first modulation matrix of the optical lens, and the second modulation matrix of the polarization element of the pixel. The first modulation matrix describes the modulation effect of the optical lens on the radiation intensity and polarization state of the incident light, and the second modulation matrix describes the modulation effect of the polarization element of the pixel on the radiation intensity and polarization state of the incident light. The incident light is incident on the polarization camera, and the state of the incident light is changed. In each state, the response value of the pixel of the polarization camera is obtained, wherein changing the state of the incident light includes changing the radiation intensity and / or polarization state and / or incident angle of the incident light. The first modulation matrix and the second modulation matrix are calibrated based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera.

2. The polarization camera calibration method according to claim 1, characterized in that, The parameters to be calibrated for the first modulation matrix include the polarization degree and polarization angle of the optical lens; The calibration of the first modulation matrix, based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, includes: The incident light is incident on the polarization camera, and the polarization angle of the incident light is changed. The response value of the pixel of the polarization camera is obtained at each polarization angle of the incident light. Based on the multiple polarization angles of the incident light and the response value of the pixel obtained at each polarization angle, the polarization degree and polarization angle of the optical lens are obtained.

3. The polarization camera calibration method according to claim 2, characterized in that, The calibration of the first modulation matrix, based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, includes: The incident light is incident on the polarization camera, the incident angle of the incident light is changed, and the polarization angle of the incident light is changed at each incident angle. The response value of the pixel of the polarization camera is obtained at each polarization angle of the incident light. Based on the multiple polarization angles of the incident light and the response value of the pixel obtained at each polarization angle, the polarization degree of the optical lens corresponding to the current incident angle is obtained. The relationship between the polarization of the optical lens and the incident angle of the incident light is obtained based on the multiple incident angles of the incident light and the polarization corresponding to each incident angle.

4. The polarization camera calibration method according to claim 1, characterized in that, The parameters to be calibrated in the second modulation matrix include the extinction ratio of the polarization element of the pixel; The calibration of the second modulation matrix, based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, includes: The incident light is incident on the polarization camera, and the polarization angle of the incident light is changed. The response value of the pixel of the polarization camera is obtained at each polarization angle of the incident light. Based on the maximum and minimum response values ​​of the obtained pixel response values, the extinction ratio of the polarization element of the pixel is obtained.

5. The polarization camera calibration method according to claim 1, characterized in that, The pixels of the detector are arranged in the form of pixel groups, and each pixel group includes at least two pixels, wherein the polarization angles of the polarization elements corresponding to the at least two pixels are different. The modulation effect of the second modulation matrix on the radiation intensity of the incident light is reflected by the relative transmittance of the pixel. The relative transmittance of the pixel describes the relationship between the response capability of the pixel and the response capability of the reference pixel in the pixel group to which the pixel belongs for incident light of the same polarization state. The calibration of the second modulation matrix, based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, includes: The incident light is incident on the polarization camera to obtain the response values ​​of at least two pixels in any pixel group; For any one of the at least two pixels, a first ratio is obtained between the response value of this pixel and the response value of a reference pixel among the at least two pixels, and the relative transmittance of this pixel is obtained based on the first ratio of this pixel.

6. The polarization camera calibration method according to claim 1, characterized in that, The modulation effect of the first modulation matrix on the radiation intensity of the incident light is reflected by the low-frequency relative transmittance of the pixel. The low-frequency relative transmittance of the pixel describes the relationship between the transmittance of the optical lens when the incident light is not perpendicular to the incident light and the transmittance of the optical lens when the incident light is perpendicular to the incident light for the same polarization state. The calibration of the first modulation matrix, based on the radiation intensity and / or polarization state and / or incident angle of the incident light and the response value of the pixel of the polarization camera, includes: The incident light is incident perpendicularly onto the polarization camera, and the response value of the pixel in the target area of ​​the detector is obtained; The incident light is incident non-perpendicularly onto the polarization camera, and the response value of the pixel in the target region of the detector is obtained; The sum of the response values ​​of the pixels in the target area when the incident light is not perpendicularly incident is obtained, and a second ratio is obtained to the sum of the response values ​​of the pixels in the target area when the incident light is perpendicularly incident. The low-frequency relative transmittance corresponding to the incident angle corresponding to the incident light being not perpendicularly incident is obtained based on the second ratio.

7. The polarization camera calibration method according to claim 6, characterized in that, The target area of ​​the detector is the central field of view of the detector, which refers to the area on the detector that is within the central field of view of the incident light.

8. The polarization camera calibration method according to claim 1, characterized in that, The parameters to be calibrated in the response model include the absolute emissivity, which describes the relationship between the pixel's response to the incident light and the radiance of the incident light. The calibration to obtain the absolute emissivity includes: The incident light is incident on the polarization camera, the radiation intensity of the incident light is changed, the response value of the pixel is obtained at each radiation intensity of the incident light, and the absolute radiometric coefficient is obtained based on the multiple radiation intensities of the incident light and the response value of the pixel obtained at each radiation intensity.

9. The polarization camera calibration method according to any one of claims 1 to 8, characterized in that, The devices used in the application include at least: A light source device for emitting raw light; A preset polarization element is disposed on the light-emitting side of the light source device to make the original light become polarized light after passing through the preset polarization element, thus forming the incident light; Changing the radiation intensity of the incident light includes: changing the radiation intensity of the original light emitted by the light source device; Or / and, changing the polarization state of the incident light includes: rotating the preset polarization element about the optical axis of the preset polarization element.

10. The polarization camera calibration method according to claim 9, characterized in that, The devices used in this application also include: A first turntable, on which the polarization camera is mounted, and the first turntable is used to drive the polarization camera to rotate; Changing the incident angle of the incident light includes: rotating the polarization camera by driving the first turntable to change the incident angle of the incident light incident on the polarization camera.

Images