Non-uniformity correction method of area array detector based on integrating sphere light source compensation

Abstract

The present invention discloses a method for correcting the non-uniformity of a planar array detector based on compensation of an integrating sphere surface light source, which belongs to the fields of detection radiometer data correction, remote sensor calibration, and optical radiation measurement, and includes: measuring the radiation intensity of all sampling points on the target surface of the planar array detector that are a certain distance away from the light outlet of the integrating sphere and parallel to the plane of the light outlet of the integrating sphere according to a radiation intensity model; substituting the radiation intensity of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the planar light source; performing non-uniformity compensation on the fitted radiation intensity of the planar light source; and using the compensated radiation intensity of the planar light source to perform non-uniformity correction on the planar array detector. This method takes into account the non-uniformity of the integrating sphere surface light source, so that in the process of correcting the non-uniformity of the planar array detector, the steps are more rigorous in detail, the data are more precise, and the correction results are more accurate.

Description

Technical Field

[0001] The invention belongs to the fields of detection radiometer data correction, remote sensor calibration and optical radiation measurement, and relates to a method for correcting the non-uniformity of a planar array detector based on integrating sphere surface light source compensation. Background Art

[0002] An area array detector is an image sensor that can capture a complete two-dimensional image at one time. It is widely used in the medical field, aviation detection field, scientific research laboratories, etc. to obtain high-quality image data. An area array detector is usually composed of multiple pixel units, which are distributed on the surface of the detector in a certain arrangement (such as a rectangular array). When the detector receives a light signal, each pixel unit converts the received light signal into an electrical signal output, which is combined into a two-dimensional image.

[0003] When each pixel of the array detector receives the same light signal, the electrical signal output by each pixel is theoretically exactly the same. In practice, due to the limitations of the manufacturing process, the non-uniformity of the detector material, the inconsistency of the amplification factor of the amplifier circuit, etc., the electrical signals output by different pixels are inconsistent, that is, non-uniformity. In practical applications, the non-uniformity correction of the array detector is crucial to improving image quality, ensuring data accuracy, and improving the performance of the detector. Therefore, researchers and engineers have been constantly exploring and optimizing non-uniformity correction technology to meet the needs and challenges of different fields. At present, the methods for correcting non-uniformity are mainly divided into three categories: the first category is to process in the integrated circuit inside the detector to compensate for the non-uniformity; the second category is to provide the detector with a uniform radiation source (integrating sphere, surface source blackbody) in the calibration of the detector die, and correct the different output values ​​of different pixels to the same value; the third category is to add a front lens to the detector, and use the image processing algorithm in the actual scene to complete the non-uniformity correction of the detector. The correction method of this application belongs to the second category.

[0004] At present, the main engineering practice is to use integrating spheres to perform non-uniformity correction on array detectors. The premise of the correction is to assume that the radiation intensity of each point of the surface light source output by the integrating sphere is exactly the same. Then, when the detector faces this surface light source, the electrical signal output by each pixel should also be consistent. For inconsistent pixels, multiplying by the corresponding proportional factor can correct them to a consistent value. However, in actual applications, there will be certain non-uniformity in the integrating sphere, which mainly comes from the following aspects: uneven reflective materials on the inner wall, uneven installation of integrating sphere components, inconsistent luminous intensity of different surface light sources in the integrating sphere, and differences caused by multiple reflections of light inside. How to test the radiation intensity of different points of the surface light source and compensate for the non-uniformity of the surface light source is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method for correcting the non-uniformity of a surface array detector based on the compensation of a surface light source using an integrating sphere. Aiming at the non-uniformity of the surface light source emitted by the integrating sphere, the method tests the radiation intensity at different points of the surface light source, uses a theoretical model to complete the non-uniformity compensation of the surface light source, and then performs non-uniformity correction on the surface array detector.

[0006] The technical solution adopted by the present invention is as follows:

[0007] The non-uniformity correction method of the area array detector based on integrating sphere light source compensation includes:

[0008] S110, measuring the radiation intensity of all sampling points on the target surface of the planar array detector that is a certain distance away from the light outlet of the integrating sphere and parallel to the plane of the light outlet of the integrating sphere according to the radiation intensity model, wherein the radiation intensity model refers to a model composed of the light outlet of the integrating sphere and the target surface of the planar array detector based on an actual measurement device;

[0009] S120, substituting the radiation intensity of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the surface light source;

[0010] S130, performing non-uniform compensation on the radiation intensity of the fitted surface light source;

[0011] S140, performing non-uniformity correction on the array detector using the compensated radiation intensity of the surface light source.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] First, in the process of non-uniformity correction of area array detectors, the non-uniformity of the integrating sphere surface light source is taken into account, and the model is used to measure, fit, interpolate and compensate for the surface light source of the integrating sphere, so that in the process of non-uniformity correction of area array detectors, the steps are more rigorous, the data are more precise, and the correction results are more accurate; second, the EMVA1288 standard requires that the distance between the target surface of the area array detector and the light outlet of the integrating sphere must be 8 times the diameter of the light outlet of the integrating sphere. After applying this method, the distance between the target surface of the area array detector and the light outlet of the integrating sphere is no longer limited in theory. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is an overall flow chart of a method for correcting non-uniformity of an area array detector based on integrating sphere surface light source compensation according to an embodiment of the present invention;

[0015] Figure 2 A schematic diagram of a radiation intensity model established according to an embodiment of the present invention;

[0016] Figure 3 A diagram of a measuring device according to an embodiment of the present invention;

[0017] Figure 4 4 is a sampling point location diagram according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The specific implementation modes of the present invention are further described in detail below with reference to the accompanying drawings.

[0019] Figure 1 FIG. 4 is an overall flow chart of a method for correcting non-uniformity of a planar array detector based on integrating sphere surface light source compensation according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0020] S110, measuring the radiation intensity of all sampling points on the target surface of the planar array detector that is a certain distance away from the light outlet of the integrating sphere and parallel to the plane of the light outlet of the integrating sphere according to the radiation intensity model, wherein the radiation intensity model refers to a model composed of the light outlet of the integrating sphere and the target surface of the planar array detector based on an actual measurement device;

[0021] S120, substituting the radiation intensity (i.e., net signal) of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the surface light source;

[0022] S130, performing non-uniform compensation on the radiation intensity of the fitted surface light source;

[0023] S140, performing non-uniformity correction on the array detector using the compensated radiation intensity of the surface light source.

[0024] Regarding the radiation intensity model mentioned in step S110, Figure 2 A 3D schematic diagram of the radiation intensity model. Figure 2 In the figure, 1 is the light outlet of the integrating sphere, and 2 is the target surface of the array detector. , the plane of the integrating sphere light outlet 1 is parallel to the target surface 2 of the array detector, is the center of the integrating sphere light outlet 1, For any point on the light outlet 1 of the integrating sphere, Point and The point distance is , the diameter of the integrating sphere light outlet 1 is , is the geometric center of the target surface 2 of the array detector, and the straight line Perpendicular to the target surface 2 of the area array detector, for The projection of the point on the target surface 2 of the area array detector, is any point on the target surface 2 of the array detector, and Point and The point distance is ,straight line With straight line The angle is Assume that the radiation intensity at each point of the integrating sphere light outlet 1 is uniform and equal to , then at any point on the target surface 2 of the array detector The radiation intensity observed at the location conforms to the following formula (1):

[0025] (1)

[0026] In the formula is the dark signal measured by the area array detector in a dark environment. In one embodiment, the diameter of the light outlet 1 of the integrating sphere used in the experiment is 100mm, The value is 800mm, and the target surface 2 of the array detector is 12mm 12mm square.

[0027] Regarding the actual measuring device mentioned in step S110, Figure 3 For the measurement device diagram, the measurement method is as follows:

[0028] Step 1: Prepare measuring instruments, including integrating sphere 3, array detector 4, slide rail 5, vertical rotating platform 6, and computer. Fix the array detector 4 on the slide rail 5, and fix the slide rail 5 on the vertical rotating platform 6. The array detector 4 can move along the slide rail 5 to simulate the change of r. The vertical rotating platform 6 can drive the array detector 4 to rotate together to simulate The target surface 2 of the area array detector used in the experiment contains 512 512 pixels. In order to ensure that more sampling points are set within a limited range, the pixel area should be as small as possible. Therefore, only the middle 4 pixels are used to measure the radiation intensity. These 4 pixels are located at the geometric center of the target surface 2 of the array detector, and the position distribution is 2 2. Therefore, before the experiment, align the center of the integrating sphere light outlet 1, the center of the area array detector target surface 2 and the center of the vertical rotating stage 6.

[0029] Step 2: Figure 4The sampling point position diagram for measurement is shown in Figure 1. Starting from 0°, a sampling radius is determined every 30°. Six sampling points are evenly distributed on each sampling radius, and adjacent sampling points (black points) on the same sampling radius are a certain distance apart. By adjusting the slide rail and the vertical rotating stage, the geometric center of the target surface of the area array detector is aligned with each sampling point in turn. The order is to first align the central sampling point, then align the first sampling point on the 0° sampling radius, then align the first sampling point on the 30° sampling radius, until aligning the first sampling point on the 330° sampling radius, then continue to align the second sampling point on the 0° sampling radius, align the second sampling point on the 30° sampling radius, and so on, until aligning the last point, which is the sixth sampling point on the 330° sampling radius.

[0030] Step 3: At the center sampling point, set the integration time of the array detector to (The signal value is about 20% of the full scale), control the array detector to collect 50 groups of light signals and 50 groups of dark signals respectively, open the shutter to measure the light signal, close the shutter to measure the dark signal, and average and subtract them to get the net signal of all pixels.

[0031] Step 4: At the center sampling point, set the integration time of the array detector to (The signal value is about 80% of the full scale), control the array detector to collect 50 groups of light signals and 50 groups of dark signals respectively, open the shutter to measure the light signal, close the shutter to measure the dark signal, average and subtract them to get the net signal of all pixels. The purpose of steps three and four is to calculate the non-uniformity correction coefficient, so that there is no need to align the geometric center of the array detector with the central sampling point again in the future to avoid introducing position errors.

[0032] Step 5: Set the integration time of the array detector to (The signal value is about 50% of the full scale). According to the order of step 2, starting from the central sampling point, at each sampling point, the array detector is controlled to collect 50 groups of light signals and 50 groups of dark signals respectively. The shutter is opened to measure the light signal, and the shutter is closed to measure the dark signal. The net signal of the central 4 pixels is obtained by averaging and subtracting. Taking one of the sampling points as an example, the calculation formulas are shown in formulas (2), (3), (4), and (5):

[0033] (2)

[0034] (3)

[0035] (4)

[0036] (5)

[0037] In the formula Respectively represent the light signal values ​​collected by the four central pixels, Represent the dark signal values ​​collected by the four central pixels, The geometric center of the target surface 2 of the array detector is located at The angle sampling radius When there are 50 sampling points, the average value of the light signal collected by the 4 central pixels is 50 times, and so on. Represents the average value of 50 collected dark signals, Represents the net signal value collected, , , corresponding to = 0mm, 10mm, 20mm, 30mm, 40mm, 50mm. In this way, the net signal value of each sampling point is obtained.

[0038] Step S120, substituting the radiation intensity (ie, net signal) of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the surface light source, including:

[0039] Step 1: Substitute the net signals of all sampling points into the radiation intensity model for fitting to obtain the radiation intensity distribution curve on each sampling radius. In one example, this step includes:

[0040] Will =800mm, =100mm, , Substitute into formula (1) and remove , we get formula (6). The premise for formula (1) is that the radiation intensity at each point of the integrating sphere light outlet 1 is uniform and equal to However, the radiation intensity at each point of the actual integrating sphere light outlet 1 is not the same, and the theoretical value needs to be corrected according to the actual observed value. Remove and add the linear factor to formula (6) and , indicating that in each There are corresponding angles and , thus we get formula (7), The net signal of all sampling points is substituted into formula (7), and the nonlinear least squares fitting method is used to estimate the value by minimizing the square error between the observed value and the model predicted value in the nonlinear model. and The most suitable value. At this time, all the parameters on the right side of formula (7) are known. =0mm~50mm is substituted into formula (6) and (7), and the obtained The radiation intensity value at each point on the angle also completes the curve fitting.

[0041] (6)

[0042] (7)

[0043] Step 2: In order to obtain the radiation intensity distribution curve of the entire surface light source on all radii, for the radii not involved in the measurement, the cubic spline interpolation method is used to obtain the corresponding net signal value. Figure 4 Take any arc 7 with a central angle of 30°, and ensure that the arc is exactly between adjacent sampling radii, extract the net signal values ​​of the sampling points corresponding to the starting point and the end point of the arc respectively, and perform cubic spline interpolation to obtain the net signal values ​​of all points on this arc. Similarly, the net signal values ​​of all other points on the arc can be obtained. All points are combined together to obtain the complete radiation intensity distribution surface of the surface light source.

[0044] S130, performing non-uniformity compensation on the radiation intensity of the fitted surface light source, including:

[0045] Taking the radiation intensity of the fitted central sampling point of the surface light source as the benchmark, the proportional relationship between the radiation intensity of other points and the radiation intensity of the central sampling point is calculated, thereby obtaining the surface light source compensation factor in polar coordinate form. In order to correspond one-to-one with the non-uniformity correction data later, it needs to be converted into rectangular coordinate form, as shown in formulas (8) and (9).

[0046] (8)

[0047] (9)

[0048] The value range is 0°~360°. The value range is 0mm~50mm, Represents the radiation intensity of the sampling point at the center of the fitted surface light source, Represents the distance between the fitted surface light source and the center sampling point And the angle is The radiation intensity at the point The surface light source compensation factor for this point is: , Represents the rectangular coordinates of this point.

[0049] S140, using the compensated surface light source radiation intensity to perform non-uniformity correction on the surface array detector, including: using a two-point correction method to perform non-uniformity correction on the surface array detector. Compared with the two-point correction method in the prior art, the present application adds a surface light source compensation factor to this method. The correction steps are as follows:

[0050] Step 1: Extract the integration time of the array detector as The net signal of all pixels measured at the central sampling point when ;

[0051] Step 2: Extract the integration time of the array detector as The net signal of all pixels measured at the central sampling point when ;

[0052] Step 3: Use the surface light source compensation factor to compensate for the surface light source, as shown in formulas (10) and (11):

[0053] (10)

[0054] (11)

[0055] , They are and During the integration time, the target surface of the array detector is located at The average of the 50 net signals measured at the pixel at , They are and During the integration time, after surface light source compensation, the target surface of the array detector is located at The pixel at the location measures the net signal value. The target surface 2 of the array detector is 12mm In the case of a 12mm square, the coordinate value cannot exceed the range of the array detector target surface 2 but must cover the four corners, so at this time The value range is mm.

[0056] Step 4: Establish a linear relationship between the image before and after non-uniformity correction, calculate the gain coefficient and the offset coefficient by the method of undetermined coefficients, and perform non-uniformity correction on the array detector, where the gain coefficient is the slope of the linear relationship and the offset coefficient is the intercept of the linear relationship.

[0057] Among them, the gain coefficient and offset coefficient of non-uniformity correction are calculated, and non-uniformity correction is performed on the array detector, as shown in formulas (12), (13), and (14). Among them, the gain coefficient is , the offset coefficient is .

[0058] (12)

[0059] (13)

[0060] (14)

[0061] is the average value of all pixels, that is, the corrected value; The coordinates are The net signal value of the pixel at , that is, the value before correction, is the integration time; The coordinates are The gain coefficient of the pixel at ; The coordinates are The offset coefficient of the pixel at ; The integration time is The average value of all pixels at time ; The integration time is The average value of all pixels at time ; The integration time is When the coordinates are The net signal value of the pixel at ; The integration time is When the coordinates are The net signal value of the pixel at ; combined with step 4, Equal to 0.5ms, Equal to 2ms, equal , equal , is equal to the average value of all pixels when the integration time is 0.5ms, It is equal to the average value of all pixels when the integration time is 2ms.

[0062] Through the above steps, the non-uniformity correction of the area array detector based on integrating sphere light source compensation is completed.

[0063] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for correcting the non-uniformity of an array detector based on integrating sphere light source compensation, characterized in that: include: S110, measuring the radiation intensity of all sampling points on the target surface of the planar array detector that is a certain distance away from the light outlet of the integrating sphere and parallel to the plane of the light outlet of the integrating sphere according to the radiation intensity model, wherein the radiation intensity model refers to a model composed of the light outlet of the integrating sphere and the target surface of the planar array detector based on an actual measurement device; S120, substituting the radiation intensity of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the surface light source; S130, performing non-uniform compensation on the radiation intensity of the fitted surface light source; S140, performing non-uniformity correction on the array detector using the compensated radiation intensity of the surface light source; The actual measuring device for obtaining the radiation intensity of all sampling points includes an array detector, a slide rail, a vertical rotating table, and an integrating sphere, wherein the array detector is fixed on the slide rail, the slide rail is fixed on the vertical rotating table, the array detector can move along the slide rail, the vertical rotating table can drive the slide rail and the array detector to rotate together, and the center of the light outlet of the integrating sphere and the center of the target surface of the array detector are aligned with the center of the vertical rotating table; The setting of the sampling points includes: starting from 0°, determining a sampling radius every 30°, and distributing 6 sampling points at equal distances on each sampling radius, and adjusting the slide rail and the vertical rotating table so that the geometric center of the target surface of the area array detector is aligned with each sampling point in turn, in the order of first aligning with the central sampling point, then aligning with the first sampling point on the 0° sampling radius, then aligning with the first sampling point on the 30° sampling radius, until aligning with the first sampling point on the 330° sampling radius, and then continuing to align with the second sampling point on the 0° sampling radius, aligning with the second sampling point on the 30° sampling radius, and so on, until aligning with the last sampling point, that is, the sixth sampling point on the 330° sampling radius; The process of obtaining the radiation intensity of all sampling points includes: Step 1: At the center sampling point, set the integration time of the array detector to The signal value of the integration time is about 20% of the full scale. The array detector is controlled to collect 50 groups of light signals and 50 groups of dark signals respectively. The light signals and dark signals are averaged and subtracted to obtain the net signal. Step 2: At the center sampling point, set the integration time of the array detector to , the signal value of the integration time is about 80% of the full scale, the array detector is controlled to collect 50 groups of light signals and 50 groups of dark signals respectively, the light signal and the dark signal are averaged and subtracted to obtain the net signal; Step 3: Set the integration time of the array detector to , the signal value of the integration time is about 50% of the full scale. Starting from the central sampling point, at each sampling point, the array detector is controlled to collect 50 groups of light signals and 50 groups of dark signals respectively. The light signal and the dark signal are averaged and subtracted to obtain the net signal; The net signal of step 1 and step 2 is used for subsequent non-uniformity correction, and the net signal of step 3 is the radiation intensity of all sampling points.

2. The method for correcting non-uniformity of an array detector based on integrating sphere light source compensation according to claim 1, characterized in that: S120, substituting the radiation intensity of all sampling points into the radiation intensity model for fitting, thereby obtaining the fitted radiation intensity of the surface light source, including: Step 1: Substitute the net signals of all sampling points into the radiation intensity model for fitting to obtain the radiation intensity distribution curve at each sampling radius; Step 2: For radii not involved in the measurement, randomly select an arc with a central angle of 30°, and ensure that the arc is exactly between adjacent sampling radii, and extract the net signal values ​​of the sampling points corresponding to the starting point and the end point of the arc respectively. Perform cubic spline interpolation on other points on the arc to obtain the net signal values ​​of all points on this arc. Similarly, the net signal values ​​of all other points on the arc are obtained. All points are combined together to obtain the complete radiation intensity distribution surface of the surface light source.

3. The method for correcting non-uniformity of an array detector based on integrating sphere light source compensation according to claim 2, characterized in that: S130, performing non-uniform compensation on the radiation intensity of the fitted surface light source, including: Step 1: Taking the radiation intensity of the fitted surface light source center sampling point as the benchmark, calculate the proportional relationship between the radiation intensity of other sampling points and the radiation intensity of the center sampling point, so as to obtain the surface light source compensation factor in polar coordinate form. , the formula is as follows: （8） Represents the angle between the line connecting the sampling point and the central sampling point and the 0° line, Represents the distance between the sampled point and the central sampling point, Represents the radiation intensity of the sampling point at the center of the fitted surface light source, Represents the distance between the fitted surface light source and the center sampling point And the angle is The radiation intensity at the sampling point; Step 2: Convert the surface light source compensation factor into rectangular coordinate form.

4. The method for correcting non-uniformity of an array detector based on integrating sphere light source compensation according to claim 3, characterized in that: S140, using the compensated surface light source radiation intensity to perform non-uniformity correction on the surface array detector, including: adding the surface light source compensation factor into the two-point correction method to perform non-uniformity correction on the surface array detector, the steps are as follows: Step 1: Extract the integration time of the array detector as The net signal measured at the central sampling point when ; Step 2: Extract the integration time of the array detector as The net signal measured at the central sampling point when ; Step 3: Use the surface light source compensation factor to compensate for the radiation intensity of the surface light source, as shown in formulas (10) and (11): （10） （11） The distance from the central sampling point And the angle is The rectangular coordinate form corresponding to the point is, , Respectively represent when the integration time is and When the target surface of the array detector is located at The average of the 50 net signals measured at the pixel at , Respectively represent when the integration time is and When the radiation intensity of the surface light source is compensated, the area on the target surface of the array detector is The net signal value of the pixel at ; Step 4: Establish a linear relationship between the image before and after non-uniformity correction, calculate the gain coefficient and the offset coefficient by the method of undetermined coefficients, and perform non-uniformity correction on the array detector, where the gain coefficient is the slope of the linear relationship and the offset coefficient is the intercept of the linear relationship.

Images