A method and system for measuring the polarization performance of a linear polarization image sensor

Abstract

The present invention provides a method and system for measuring the polarization performance of a linear polarization image sensor, including: respectively under bright-field conditions and dark-field conditions, at each set exposure amount, controlling a turntable to rotate a high-performance linear polarizer, and using a linear polarization image sensor without an optical lens to take a plurality of pictures; selecting the central region of the pictures, obtaining gray-scale response values in different polarization directions under a set exposure amount, and calculating the extinction ratio, angle calibration value, calibration parameter matrix, and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of a curve function; evaluating the polarization performance of the linear polarization image sensor by measuring the extinction ratio, angle calibration value, calibration parameter matrix, and quantum efficiency in each polarization direction. The present invention solves the defect that it is difficult to accurately measure the polarization performance of existing linear polarization image sensors, so as to achieve accurate evaluation of the polarization performance of linear polarization image sensors.

Description

Technical Field

[0001] The present invention relates to the technical field of polarization imaging, and particularly to a polarization performance measurement device and method for a linear polarization image sensor. Background Art

[0002] Currently, there is no unified standard for testing linear polarization image sensors. The testing standards for image sensors at home and abroad are the camera and image sensor testing standards formulated by the European Machine Vision Association: EMVA Standard1288. This standard is applicable to monochromatic or color digital cameras with linear characteristics, as well as analog cameras and image sensors with image acquisition cards. Currently, most of the commercially available integrated split focal plane polarization image sensors are linear polarization sensors. The EMVA1288 standard has a wider scope of application. It is proposed to improve the EMVA 1288 standard to adapt to the measurement of the response characteristics of the pixels of the integrated polarization sensor.

[0003] The existing polarization sensor performance evaluation and analysis noise models are usually a combined model of Gaussian noise and Poisson noise, which has a large error from the actual measurement values. This is mainly because the performance parameter testing of linear polarization sensors is not yet perfect, and quantum efficiency, extinction ratio, angular error, and uniformity are not taken into account. Therefore, to obtain a complete error model closer to the true value, it is necessary to measure and substitute various performance parameters of the linear polarization image sensor. Summary of the Invention

[0004] The present invention provides a polarization performance testing device and method for a linear polarization image sensor, which is used to solve the polarization performance parameters that are difficult to accurately measure for a linear polarization image sensor, so as to achieve an accurate evaluation of the polarization performance of the linear polarization image sensor.

[0005] The present invention provides a polarization performance measurement method for a linear polarization image sensor, including:

[0006] Under bright field conditions and dark field conditions respectively, at each set exposure amount, rotate the linear polarizer, and take a plurality of pictures through the linear polarization image sensor;

[0007] Select the central region of the picture to obtain the gray response values of pixels in different polarization directions, and obtain the extinction ratio, angle calibration value, correction parameter matrix, and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of the curve function;

[0008] Evaluate the polarization performance of the linear polarization image sensor by measuring the extinction ratio, angle calibration value, correction parameter matrix, and quantum efficiency in each polarization direction.

[0009] A method for measuring the polarization performance of a linear polarization image sensor provided by the present invention, wherein the central region of the selected picture is obtained, and the gray-scale response values of pixels in different polarization directions are obtained. The extinction ratio, angle calibration value, correction parameter matrix, and quantum efficiency in each polarization direction are obtained by fitting functions or substituting into physical formulas of curve functions, specifically including:

[0010] In bright-field and dark-field environments, at a certain relative rotation angle of the turntable, use an optical-lensless device equipped with a linear polarization image sensor to take a number of pictures, observe the difference in gray-scale values in the same polarization direction, and adjust the fixed angle of the linear polarization sensor;

[0011] In a bright-field environment, set different equally spaced exposure times for the linear polarization image sensor, rotate the linear polarizer, and take M pictures for N cycles. Select the central region of the picture that contains pixels in different polarization directions;

[0012] At equally spaced exposure times and wavelengths, rotate the turntable to obtain a sequence of the mean gray-scale responses of pixels in each polarization direction varying with the relative rotation angle, and perform Fourier fitting to obtain a fitting function, where the mean value of the peak of the fitting curve is μ y ;

[0013] Set at a certain specific wavelength and exposure amount, rotate the turntable to obtain a first set of sequences of the mean gray-scale responses of pixels in each polarization direction varying with the relative rotation angle, perform Fourier fitting to obtain a fitting function, calculate the phase difference between the functions in different polarization directions, and obtain the angle calibration value;

[0014] Set the same wavelength and exposure amount, rotate the turntable to obtain a sequence of the gray-scale response values of pixels in each polarization direction varying with the relative rotation angle. Select the sequences of the gray-scale response values of pixels in each polarization direction in the central region varying with the rotation angle for fitting, and calculate the average value of the peak and valley of the function. The ratio of the peak to the valley is the extinction ratio of pixels in each polarization direction;

[0015] In a dark-field environment, control the exposure time to change equally spaced with the bright-field condition to obtain the dark signal output values at different exposure times t exp Finally, obtain the average value sequence μ y.dark , calculate the variance of the gray-scale values in bright-field and dark-field environments, and fit the variance varying with the average gray-scale response (μ y -μ y.dark ) to obtain a first linear curve. The slope of the first linear curve is the overall system gain K;

[0016] Under the same bright-field conditions, set the exposure time and wavelength at the same equal intervals. Place the calibrated photodiode at the same position as the linear polarization image sensor, rotate the linear polarizer, and calculate the average value E of the irradiance per unit area;

[0017] Fit the exposure amount (μ y -μ y.dark ) of the linear polarization image sensor as a function of the second linear curve that changes with the exposure time t exp Obtain the slope of the second linear curve, substitute it into the physical formula of the slope, and obtain the quantum efficiency η;

[0018] Convert the filter of different wavelengths to obtain different quantum efficiencies, and draw the curve of the relationship between the wavelength λ and η(λ).

[0019] According to a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention, the method for obtaining the calibration parameter matrix from the picture specifically includes:

[0020] Under bright-field conditions, set multiple exposure times and wavelengths at equal intervals, rotate the linear polarizer, and for each rotation angle, the linear polarization image sensor takes multiple pictures, and select the central area of the picture;

[0021] At a certain specific exposure time and a certain specific wavelength, perform a least squares fit on the measured values and ideal values of the gray-scale responses of all pixels in each polarization direction to obtain the uniformity calibration parameter matrix and the dark noise matrix of all superpixels;

[0022] Among them, regard the gray-scale response value obtained by the calibrated photodiode as the ideal light intensity value.

[0023] According to a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention, the method for setting at a certain specific wavelength and exposure amount, rotating the turntable, and obtaining the first set of sequences of the mean gray-scale responses of pixels in each polarization direction changing with the relative rotation angle specifically includes:

[0024] Control the exposure amount by changing the illumination intensity of the integrating sphere light source or adjusting the exposure time of the linear polarization image sensor;

[0025] Rotate the high-performance linear polarizer, and at a certain specific exposure amount, obtain the sequence of the gray-scale response values in each polarization direction changing with the relative rotation angle;

[0026] For each polarization direction, obtain the first set of sequences of the mean gray-scale responses of pixels in each polarization direction changing with the relative rotation angle.

[0027] According to a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention, the method for obtaining the angle calibration value and the extinction ratio is:

[0028] Perform Fourier fitting on the first set of sequences to obtain the fitting functions of the gray response values of the pixels in each polarization direction varying with the relative rotation angle, and calculate the peak and valley values of the fitting functions;

[0029] Select the sequences of the gray response values of the pixels in each polarization direction varying with the rotation angle for Fourier fitting, calculate the phase difference between the fitting functions in different polarization directions, and generate the angle calibration value;

[0030] Calculate the average value of the peak and valley values of the fitting function, and the ratio of the peak value to the valley value is the extinction ratio of the polarization array.

[0031] According to a performance measurement method of a linear polarization image sensor provided by the present invention, fit the second linear curve of the exposure amount of the polarization image sensor varying with the exposure time, and obtain the quantum efficiency according to the slope of the second linear curve, specifically including:

[0032] According to the exposure amount (μ y -μ y.dark ) of the linear polarization image sensor varying with the exposure time t exp to fit and generate the second linear curve equation,

[0033] Substitute the total system gain K, wavelength λ, average illumination intensity E, Planck constant h, unit area A, and the speed of light c into the slope of the second linear curve equation to obtain the quantum efficiency η.

[0034] The present invention also provides a polarization performance measurement system for a linear polarization image sensor. The system includes:

[0035] An image acquisition module, which rotates a linear polarizer at each set exposure amount under bright field conditions and dark field conditions respectively, and takes a plurality of pictures through the linear polarization image sensor;

[0036] A parameter calculation module, which selects the central area of the picture to obtain the gray response values of the pixels in different polarization directions, and obtains the extinction ratio, angle calibration value, calibration parameter matrix, and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of the curve function;

[0037] A polarization performance evaluation module, which is used to evaluate the polarization performance of the linear polarization image sensor through the extinction ratio, angle calibration value, calibration parameter matrix, and quantum efficiency in each polarization direction.

[0038] A method and system for measuring the polarization performance of a linear polarization image sensor provided by the present invention rotate a turntable in a bright-field environment and a dark-field environment, use the linear polarization image sensor to take multiple pictures, obtain a fitting function through fitting, calculate the phase difference and peak-to-valley value, or substitute into a physical relationship formula to obtain the extinction ratio, angle calibration value, correction parameter matrix, and quantum efficiency in each polarization direction, and measure the polarization performance of the linear polarization image sensor. Accurate measurement of relevant data is achieved, and the performance of the linear polarization image sensor can be evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0040] Figure 1 is one of the flow diagrams of a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0041] Figure 2 is the second of the flow diagrams of a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0042] Figure 3 is the third of the flow diagrams of a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0043] Figure 4 is the fourth of the flow diagrams of a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0044] Figure 5 is the fifth of the flow diagrams of a method for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0045] Figure 6 is a schematic diagram of the light propagation of a device for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0046] Figure 7 is a schematic diagram of the module connection of a system for measuring the polarization performance of a linear polarization image sensor provided by the present invention;

[0047] Reference numerals:

[0048] 110: Image acquisition module; 120: Parameter calculation module; 130: Polarization performance evaluation module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] The following will describe Figures 1-5 a method for measuring the polarization performance of a linear polarization image sensor according to the present invention, including:

[0051] S100. Respectively under bright field conditions and dark field conditions, set equally spaced exposure times and wavelengths, rotate the linear polarizer, and take multiple pictures through the linear polarization image sensor;

[0052] S200. Select the central area of the picture, obtain the sequence of the gray response values of each polarization direction changing with the relative rotation angle, and through the fitted function, calculate the phase difference and the peak-to-valley value, or substitute into the physical relationship formula to obtain the extinction ratio, the angle calibration value, the calibration parameter matrix, and the quantum efficiency of each polarization direction;

[0053] S300. Evaluate the polarization performance of the linear polarization image sensor through the extinction ratio, the angle calibration value, the calibration parameter matrix, and the quantum efficiency of each polarization direction.

[0054] Referring to Figure 6 , in order to obtain the images captured by the linear polarization image sensor, an integrating sphere light source, a filter wheel, a collimator, a linear polarizer with high extinction ratio and transmittance, a turntable, a calibrated photodiode, a lensless device equipped with a linear polarization image sensor, and a displacement platform are set in the present invention. The path of the light is as follows: the parallel and uniform light emitted by the integrating sphere light source passes through the filter wheel, the collimator, and the rotatable high-performance linear polarizer fixed on the turntable in sequence, and finally reaches the photosensitive surface of the linear polarization image sensor fixed by the displacement platform or the bracket or the calibrated photodiode. Multiple images are taken through the linear polarization image sensor.

[0055] And by adjusting the parameters of the parallel and uniform light source emitted by the integrating sphere light source and adjusting the filter wheel, the wavelength of the parallel and uniform light incident on the photosensitive surface of the linear polarization image sensor can be changed.

[0056] Obtain the polarization gray response values in different directions from the pictures, and obtain the extinction ratio, the angle calibration value, the calibration parameter matrix, and the quantum efficiency of each polarization direction through the fitting function or substituting into the relationship equation, specifically including:

[0057] S201. In bright-field and dark-field environments, at a certain relative rotation angle of the turntable, use a lensless device equipped with a linear polarization image sensor to take a number of pictures, observe the difference in gray values in the same polarization direction, and adjust the fixed angle of the linear polarization sensor;

[0058] S202. In the bright-field environment, set different equally spaced exposure times for the linear polarization image sensor, rotate the linear polarizer, take M pictures for N cycles, and select pixels with different polarization directions in the central area of the pictures;

[0059] In the present invention, set the exposure times to be 1s, 2s,..., 10s respectively. At these 10 exposure times, measure the pixel response values changing with the rotation of the turntable under bright-field and dark-field conditions respectively, where the turntable drives the rotatable high-performance linear polarizer to rotate clockwise with an increment of 1° per step, and rotates a total of 360 times;

[0060] S203. In the bright-field environment, at equally spaced exposure times and wavelengths, rotate the turntable to obtain a sequence of the mean gray-scale responses of pixels in each polarization direction changing with the relative rotation angle, and perform Fourier fitting to obtain a fitting function, where the mean of the peak values of the fitting curve is μ y ;

[0061] S204. In the bright-field environment, set a certain specific wavelength and exposure amount, rotate the turntable to obtain a first set of sequences of the mean gray-scale responses of pixels in each polarization direction changing with the relative rotation angle, perform Fourier fitting to obtain a fitting function, calculate the phase difference between the functions in different polarization directions, and obtain an angle calibration value;

[0062] S205. Set the same wavelength and exposure time, rotate the turntable to obtain a sequence of the gray-scale response values of all pixels in each polarization direction changing with the relative rotation angle, select the sequences of the gray-scale responses of each polarization direction in the central area changing with the rotation angle for fitting, and calculate the average of the peak and valley values of the function. The ratio of the peak value to the valley value is the extinction ratio of the pixels in each polarization direction;

[0063] S206. In the dark-field environment, control the exposure time to change equally spaced as in the bright-field condition to obtain the dark signal output values at different exposure times t exp Finally, obtain the average value sequence μ y.dark , calculate the variance of the gray values in the bright-field and dark-field environments, and fit the variance changing with the average gray-scale response (μ y -μ y.dark ) to obtain a first linear curve. The slope of the first linear curve is the overall system gain K;

[0064] S207. Under the same bright-field conditions, set the exposure time and wavelength at the same equal intervals. Place the calibrated photodiode at the same position as the linear polarization image sensor, rotate the linear polarizer, and calculate the average value E of the irradiance per unit area;

[0065] S208. Fit the second linear curve function of the exposure amount (μ y -μ y.dark ) varying with the exposure time t exp . Obtain the slope of the second linear curve, substitute it into the slope physical formula, and obtain the quantum efficiency η;

[0066] S209. Switch the filter of different wavelengths to obtain different quantum efficiencies, and draw the curve of the relationship between the wavelength λ and η(λ).

[0067] Obtain the uniformity correction parameter matrix, specifically including:

[0068] S301. Under the bright-field conditions, at a certain specific exposure time and a certain specific wavelength, rotate the linear polarizer. For each rotation angle, the linear polarization image sensor takes multiple pictures, and select the central area of the pictures;

[0069] S302. Perform the least-squares fitting on the measured values and ideal values of the gray-scale responses of all pixels in each polarization direction to obtain the uniformity correction parameter matrix and the dark noise matrix of all superpixels;

[0070] Among them, the ideal light intensity value is obtained through the gray-scale response value of the calibrated photodiode.

[0071] By obtaining the extinction ratio, the angle calibration value, the correction parameter matrix, and the quantum efficiency in each polarization direction, the polarization performance of the linear polarization image sensor can be evaluated.

[0072] Under the bright-field environment, set at a certain specific wavelength and exposure amount, rotate the turntable to obtain the first set of sequences of the average gray-scale responses of the pixels in each polarization direction varying with the relative rotation angle, specifically including:

[0073] S2021. Control the exposure amount by changing the illumination intensity of the integrating sphere light source or adjusting the exposure time of the linear polarization image sensor;

[0074] S2022. Rotate the high-performance linear polarizer, and at a certain specific exposure amount, obtain the sequence of the average gray-scale responses of each polarization direction varying with the relative rotation angle;

[0075] S2023. For each polarization direction, obtain the first set of sequences of the average gray-scale responses of the pixels in each polarization direction varying with the relative rotation angle.

[0076] In the bright-field environment, by changing the light intensity of the integrating sphere light source or setting the exposure time of the camera to control the exposure amount, at a certain specific exposure amount, rotate the turntable and take M pictures for N cycles. There are 4 polarization directions in the selected area of the pictures. Let I be the number of pixels in a certain polarization direction. Denote the nth (1 ≤ n ≤ N) picture in this polarization direction and the response value of the ith (1 ≤ i ≤ I) pixel. Then, the average value μ of the pixel responses in this polarization direction at this relative rotation angle is: The gray response values of all pixels are

[0077] In the present invention, select an area in the picture where the number of pixels in each polarization direction is 30000, and calculate the average value of the pixel responses in each polarization direction when the rotatable high-performance linear polarizer rotates to this position. Denote the response value of the ith (1 ≤ i ≤ 30000) pixel in a certain polarization direction as x i , then the average value μ of the pixel responses of this polarization pixel when the rotatable high-performance linear polarizer rotates to this angle is:

[0078] For each angle of rotation of the rotatable high-performance linear polarizer, the above method is adopted, and finally a sequence of the average values of the pixel responses containing 360 points in each polarization direction at different exposure times is obtained.

[0079] Set different exposure amounts and wavelengths at equal intervals, and adopt the above method to obtain a sequence of the gray response means of the pixels in each polarization direction changing with the relative rotation angle; in the bright-field environment, at a certain specific exposure amount and wavelength, the first group of sequences of the gray response means of the pixels in each polarization direction changing with the relative rotation angle.

[0080] Perform Fourier fitting on the first group of sequences to obtain a fitting function. Select the pixels in the central area of the image to obtain the fitting function of its gray response value changing with the relative rotation angle, and obtain the phase difference between the fitting functions in each polarization direction, that is, the angle calibration value, and the ratio of the average value of the peak and valley values of the fitting curve, that is, the extinction ratio.

[0081] In the dark-field environment, control the equal-interval changes of the exposure time and bright-field conditions, measure the gray values of the linear polarization image sensor N times and calculate the sequence of the average gray value responses; calculate the variance of the gray values under the bright-field conditions and dark-field conditions. For example, when the number of acquisition times N = 2, the pixels in the same polarization direction are combined into a pixel map, and its gray value variance is Fit the gray value variance at different rotation angles changing with the average gray value (μ y -μ y.dark ) to obtain a second linear curve, and its slope is the overall system gain K.

[0082] Fix a calibrated photodiode at the same position of the above-mentioned experimental linear polarization image sensor, set the light intensity of the light emitted from the integrating sphere to be the same as that in the dark field environment, set the exposure time to change at equal intervals in the dark field environment, rotate the turntable under the same bright field environment, measure N cycles, and calculate the average light intensity E per unit area A.

[0083] Fit the exposure amount (μ y -μ y.dark ) of the linear polarization image sensor as a function of the exposure time t exp to obtain the second linear curve. Calculate the quantum efficiency according to the slope of the second linear curve, specifically including:

[0084] S2061. Fit the second linear curve equation of the exposure amount (μ y -μ y.dark ) of the linear polarization image sensor as a function of the exposure time t exp . The slope of the curve is where substituting the total system gain K, wavelength λ, average light intensity E, Planck's constant h, unit area A, and the speed of light c can obtain the quantum efficiency η. By changing the filter of different wavelengths, different quantum efficiencies can be obtained, and thus the curve of the change relationship between the wavelength λ and η(λ) can be drawn.

[0085] Through the linear curve function of the difference between the average pixel gray response values of each polarization direction in the bright and dark fields and the average exposure time t exp , obtain its slope, substitute each known and calculated constant to solve for η, measure the function curve of η as a function of wavelength under different wavelength conditions, solve the problem of measuring the quantum efficiency of each polarization direction in the linear polarization image sensor, and also obtain the extinction ratio, angle calibration value, and calibration parameter matrix, which is convenient for evaluating the performance of the linear polarization image sensor.

[0086] Reference Figure 7 , the present invention also discloses a polarization performance measurement system for a linear polarization image sensor, and the system includes:

[0087] An image acquisition module 110, which is used to rotate the linear polarizer respectively under bright field conditions and dark field conditions. Each time it rotates, use the linear polarization image sensor to take multiple pictures;

[0088] A parameter calculation module 120, which is used to select the pixel in the central area of the image to obtain the sequence of the gray response values of different polarization directions changing with the relative rotation angle, calculate the phase difference and peak-to-valley value through the fitting function, or obtain the extinction ratio, angle calibration value, calibration parameter matrix, and the quantum efficiency of each polarization direction by substituting the physical relationship formula;

[0089] The polarization performance evaluation module 130 is used to evaluate the polarization performance of the linear polarization image sensor based on the extinction ratio, the angle calibration value, the correction parameter matrix, and the quantum efficiency in each polarization direction.

[0090] The image acquisition module 110, under bright field conditions, sets equally spaced exposure times and wavelengths, rotates the linear polarizer, and each time it rotates, the linear polarization image sensor takes multiple pictures;

[0091] The exposure amount is changed by changing the illumination intensity of the light source or adjusting the exposure time;

[0092] The linear polarizer is rotated, and at a certain specific exposure amount and wavelength, a first set of sequences of the mean gray response of pixels in each polarization direction changing with the relative rotation angle is obtained.

[0093] The parameter calculation module 120 fits the first set of sequences by Fourier fitting method to obtain the fitting function of the mean gray response of pixels in each polarization direction changing with the relative rotation angle, and calculates the phase difference between the fitting functions in different polarization directions, that is, the angle calibration value;

[0094] The mean values of the peak and valley of the fitting function in each polarization direction are calculated, and the ratio of the peak to the valley is the extinction ratio of the polarization array.

[0095] According to the exposure amount (μ y -μ y.dark ) of the linear polarization image sensor changing with the exposure time t exp a second linear curve equation is fitted and generated,

[0096] The total system gain, wavelength, mean illumination intensity, unit area, Planck's constant, and the speed of light are substituted into the slope of the second linear curve equation to obtain the quantum efficiency.

[0097] The exposure amount (μ y -μ y.dark ) of the linear polarization image sensor changing with the exposure time t exp is fitted to generate a second linear curve equation, and the slope of the curve is where substituting the total system gain K, wavelength λ, mean illumination intensity E, Planck's constant h, unit area A, and the speed of light c can obtain the quantum efficiency η. By changing the filter with different wavelengths, different quantum efficiencies can be obtained, and thus the change relationship curve between the wavelength λ and μ(λ) can be drawn.

[0098] The polarization performance evaluation module 130 measures the polarization performance of the linear polarization image sensor by rotating the turntable under bright-field and dark-field conditions, acquiring multiple pictures captured by the linear polarization image sensor, and obtaining the extinction ratio, angle calibration value, calibration parameter matrix, and quantum efficiency in each polarization direction, thus achieving accurate measurement of relevant data and enabling accurate evaluation of the performance of the linear polarization image sensor.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or equivalently replace some of the technical features. However, such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for measuring the polarization performance of a linear polarization image sensor, characterized in that, it includes: Under bright field conditions and dark field conditions respectively, at each set exposure amount, rotate the linear polarizer, and take several pictures through the polarization image sensor; Select the central area of the picture to obtain the gray response values of pixels in different polarization directions, and obtain the extinction ratio, angle calibration value, calibration parameter matrix and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of the curve function; Evaluate the polarization performance of the linear polarization image sensor by measuring the extinction ratio, angle calibration value, calibration parameter matrix and quantum efficiency in each polarization direction; Among them, selecting the central area of the picture to obtain the gray response values of pixels in different polarization directions, and obtaining the extinction ratio, angle calibration value, calibration parameter matrix and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of the curve function specifically includes: In bright field and dark field environments, at a certain relative rotation angle of the turntable, use an optical lensless device equipped with a linear polarization image sensor to take several pictures, and observe the difference in gray values in the same polarization direction to adjust the fixed angle of the polarization sensor; In a bright field environment, set different equally spaced exposure times for the linear polarization image sensor, rotate the linear polarizer, and take M pictures in N cycles. Select the central area of the picture to contain pixels in different polarization directions; At equal-spacing exposure time and wavelength, rotate the turntable to obtain a sequence of the mean gray response of pixels in each polarization direction varying with the relative rotation angle, and perform Fourier fitting to obtain a fitting function, where the mean of the peak values of the fitting curve is μ y ; Set at a certain specific wavelength and exposure amount, rotate the turntable to obtain the first set of sequences of the average gray response of pixels in each polarization direction changing with the relative rotation angle, perform Fourier fitting to obtain the fitting function, calculate the phase difference between the functions in different polarization directions, and obtain the angle calibration value; Set the same wavelength and exposure amount, rotate the turntable to obtain the sequence of the gray response values of pixels in each polarization direction changing with the relative rotation angle. Select the sequence of the gray response values of pixels in each polarization direction in the central area changing with the rotation angle for fitting, and calculate the average value of the peak and valley values of the function, calculate the ratio of the peak value to the valley value, and obtain the extinction ratio of pixels in each polarization direction; Under dark field conditions, the exposure time is controlled to change at equal intervals from the bright field conditions, and different exposure times t are obtained. exp The dark signal output values are obtained, and finally the average value sequence μ is calculated. y.dark The variance of the gray values between the bright field and the dark field is calculated, and the variance of the gray values is fitted. The first linear curve of the change with the average gray response value (μ y - μ y.dark ) is obtained. The slope of the first linear curve is the overall system gain K. Under the same bright field conditions, set the same equally spaced exposure time and wavelength, place the calibrated photodiode at the same position as the linear polarization image sensor, rotate the linear polarizer, and calculate the average value E of the irradiance per unit area; Fit out the second linear curve function of the exposure amount (μ y -μ y.dark ) of the linearly polarized image sensor varying with the exposure time t exp ; obtain the slope of the second linear curve, substitute it into the physical formula of the slope, and obtain the quantum efficiency η; Switch the filter of different wavelengths to obtain different quantum efficiencies, and draw the change relationship curve of wavelength λ and η(λ).

2. The method for measuring the polarization performance of a linear polarization image sensor according to claim 1, characterized in that, The selection of the central area of the picture to obtain the gray response values of pixels in different polarization directions, and obtaining the extinction ratio, angle calibration value, calibration parameter matrix and quantum efficiency in each polarization direction through a fitting function or by substituting into the physical formula of the curve function further includes: Under bright field conditions, set multiple equally spaced exposure times and wavelengths, rotate the linear polarizer, and for each rotation angle, the polarization image sensor takes multiple pictures and selects the central area of the picture; At a specific exposure time and a specific wavelength, the measured gray-scale response values of all pixels in each polarization direction are least-squares fitted with the ideal values to obtain the uniformity correction parameter matrix and the dark noise matrix of all superpixels, where a superpixel is composed of pixels with 4 different polarization directions; Among them, the gray-scale response value obtained by the calibrated photodiode is regarded as the ideal light intensity value.

3. The polarization performance measurement method of the linear polarization image sensor according to claim 1, characterized in that, at a specific wavelength and exposure amount, the turntable is rotated to obtain a first set of sequences of the average gray-scale response of pixels in each polarization direction changing with the relative rotation angle, specifically including: controlling the exposure amount by changing the illumination intensity of the integrating sphere light source or adjusting the exposure time of the linear polarization image sensor; rotating the high-performance linear polarizer, and obtaining a sequence of the average gray-scale response of each polarization direction changing with the relative rotation angle at a specific exposure amount; for each polarization direction, obtaining a first set of sequences of the average gray-scale response of pixels in each polarization direction changing with the relative rotation angle.

4. The polarization performance measurement method of the linear polarization image sensor according to claim 1, characterized in that, the method for obtaining the angle calibration value and the extinction ratio is: performing Fourier fitting on the first set of sequences to obtain a fitting function of the gray-scale response value of pixels in each polarization direction changing with the relative rotation angle, and calculating the peak value and the valley value of the fitting function; selecting the sequences of the gray-scale response values of pixels in each polarization direction changing with the rotation angle for Fourier fitting, calculating the phase difference between the fitting functions of different polarization directions, and generating an angle calibration value; calculating the average value of the peak value and the valley value of the fitting function, and the ratio of the peak value to the valley value is the extinction ratio of the polarization array.

5. The polarization performance measurement method of the linear polarization image sensor according to claim 1, characterized in that, fitting a second linear curve of the exposure amount of the polarization image sensor changing with the exposure time, and obtaining the quantum efficiency according to the slope of the second linear curve, specifically including: According to the exposure amount (μ y -μ y.dark ) of the polarization image sensor varying with the exposure time t exp to fit and generate the second linear curve equation, Substitute the total system gain K, wavelength λ, mean illumination intensity E, Planck's constant h, unit area A, and speed of light c into the slope of the second linear curve equation to obtain the quantum efficiency η. ​ 6. A polarization performance measurement system for a linear polarization image sensor, characterized in that, the system includes: a picture acquisition module, which rotates the linear polarizer at each set exposure amount under bright-field conditions and dark-field conditions respectively, and takes a number of pictures through the polarization image sensor; a parameter calculation module, which selects the central area of the picture to obtain the gray-scale response values of pixels in different polarization directions, and obtains the extinction ratio, the angle calibration value, the correction parameter matrix and the quantum efficiency of each polarization direction through a fitting function or substituting into the physical formula of the curve function; a polarization performance evaluation module, which is used to evaluate the polarization performance of the linear polarization image sensor through the extinction ratio, the angle calibration value, the correction parameter matrix and the quantum efficiency of each polarization direction; wherein, selecting the central area of the picture to obtain the gray-scale response values of pixels in different polarization directions, and obtaining the extinction ratio, the angle calibration value, the correction parameter matrix and the quantum efficiency of each polarization direction through a fitting function or substituting into the physical formula of the curve function, specifically including: In bright-field and dark-field environments, at a certain relative rotation angle of the turntable, several pictures are taken using a lensless device equipped with a linear polarization image sensor to observe the difference in gray values in the same polarization direction, so as to adjust the fixed angle of the polarization sensor; In a bright-field environment, set different equally spaced exposure times for the linear polarization image sensor, rotate the linear polarizer, and take M pictures for N cycles. Select pixels with different polarization directions in the central area of the pictures; Under equal-spacing exposure time and wavelength, rotate the turntable to obtain a sequence of the mean gray-scale responses of pixels in each polarization direction varying with the relative rotation angle, and perform Fourier fitting to obtain a fitting function, where the mean of the peak values of the fitting curve is μ y ; Set at a certain specific wavelength and exposure amount, rotate the turntable to obtain the first set of sequences of the mean gray response of pixels in each polarization direction varying with the relative rotation angle, perform Fourier fitting to obtain the fitting function, calculate the phase difference between the functions in different polarization directions, and obtain the angle calibration value; Set the same wavelength and exposure amount, rotate the turntable to obtain the sequences of the gray response values of pixels in each polarization direction varying with the relative rotation angle. Select the sequences of the gray response values of pixels in each polarization direction in the central area varying with the rotation angle for fitting, calculate the average value of the peak and valley values of the function, and calculate the ratio of the peak value to the valley value to obtain the extinction ratio of pixels in each polarization direction; Under dark field conditions, the exposure time is controlled to change at equal intervals from the bright field conditions, and different exposure times t are obtained. exp The dark signal output values are obtained, and finally the average value sequence μ is calculated. y.dark The variance of the gray values in the bright field and dark field conditions is calculated, and the variance of the gray values is fitted. The first linear curve showing the change of the variance of the gray values with the average gray response (μ y - μ y.dark ) is obtained, and the slope of the first linear curve is the overall system gain K. Under the same bright-field conditions, set the same equally spaced exposure time and wavelength, place the calibrated photodiode at the same position as the linear polarization image sensor, rotate the linear polarizer, and calculate the average value E of the irradiance per unit area; Fitting out the second linear curve function of the exposure amount (μ y -μ y.dark ) of the linearly polarized image sensor varying with the exposure time t exp to obtain the slope of the second linear curve, substituting it into the physical formula of the slope, and obtaining the quantum efficiency η; Switch filter plates of different wavelengths to obtain different quantum efficiencies, and draw the curve of the variation relationship between the wavelength λ and η(λ).

Images