Non-uniformity correction method for large-area-array infrared detector

Abstract

The invention relates to the technical field of detector correction, and discloses a large-area-array infrared detector heterogeneity correction method. Comprising a reference and saturation voltage determination step, an integral time point selection step, a diaphragm parameter setting step, a dual-temperature calibration data acquisition step, a standard response function construction step, a pixel response inverse function solving step and a signal matrix correction step. According to the method, through combination of the conformal interpolation method, the dual-temperature calibration and the logarithm equal interval integral time point design, the problem of non-uniformity correction of the large-scale infrared detector is solved, and reliable technical support is provided for a high-precision infrared imaging system.

Description

Technical Field

[0001] The present application relates to the technical field of detector correction, and in particular to a method for correcting non-uniformity of large-array infrared detectors. Background Art

[0002] Infrared focal plane array (FPA) detectors are widely used in thermal imaging, remote sensing, security, and other fields, and their performance directly impacts system imaging quality. However, as the size of the detector array increases (e.g., from hundreds of pixels to tens of thousands), non-uniformity issues (including spatial and temporal non-uniformity, nonlinear response variations, and wavelength response variations) have an increasingly significant impact on imaging quality.

[0003] In existing technology, the non-uniformity testing of infrared detectors in China is mainly based on the GB / T17444 standard. However, this method only tests the device output signal and does not involve non-uniformity correction. Traditional correction methods (such as calibration correction and adaptive correction) have the following limitations:

[0004] Risk of overcorrection: Directly correcting non-uniformity by adjusting system parameters (such as gain and offset) may mask device process or design defects and fail to accurately reflect the device's true non-uniformity.

[0005] Limited dynamic range: Existing methods have difficulty taking into account the nonlinear response characteristics of the detector within a wide dynamic range, resulting in insufficient uniformity and dynamic range of the corrected image.

[0006] Insufficient applicability: The non-uniformity of large-scale array detectors has complex spatial distribution characteristics, and traditional single-point or local correction methods are difficult to meet high-precision requirements.

[0007] In addition, during the testing process, the infrared detector needs to ensure that the radiation source (such as a point source blackbody) covers the entire field of view, but the existing technology does not clearly define the correlation between aperture parameters and non-uniformity correction, resulting in limited correction accuracy.

[0008] Therefore, a new infrared detector non-uniformity correction technology is urgently needed to improve the accuracy of the correction results. Summary of the Invention

[0009] The purpose of this application is to provide a method for correcting non-uniformity of a large-array infrared detector to solve the technical problems raised in the above-mentioned background technology.

[0010] To achieve the above objectives, the present application discloses the following technical solution: a method for correcting non-uniformity of a large array infrared detector, the method comprising the following steps:

[0011] Determination of reference and saturation voltage: Place the detector in a completely dark environment and obtain the mean signal voltage as the reference voltage. Place the detector in front of a radiation source and adjust the incident radiation intensity so that the mean signal voltage reaches a minimum value, which is defined as the saturation voltage.

[0012] Integration time point selection: According to the actual working integration time range of the detector [t m ,t n ], select N integration time points t1, t2,…, t N , and each integral time point satisfies log(t n )-log(t n -1)=log(t n +1)-log(t n );

[0013] Aperture parameter setting: Adjust the integration time to the maximum integration time point t N , change the aperture diameter D and the distance L from the detector to the aperture, and make the mean value of the signal voltage close to the saturation voltage under the condition of paraxial approximation;

[0014] Dual-temperature calibration data acquisition: Keep the aperture diameter D and the distance from the detector to the aperture L unchanged, and calculate the integral time points t1, t2, ..., t N Change the integration time in sequence and record the signal voltage value of each effective pixel at temperature T1 and T2;

[0015] Standard response function construction: based on the integration time points t1, t2, ..., t N The corresponding effective pixel signal voltage mean V t1 ,V t2 ,…,V tN , the standard response function is generated using the shape-preserving piecewise cubic interpolation method;

[0016] Solve the inverse function of pixel response: For the pixel (i, j) in the i-th row and j-th column, based on the integration time t ij =t1,t2,…,t N The signal voltage value under The inverse function V′=F of the response function is solved by using the shape-preserving piecewise cubic interpolation method. V (V ij );

[0017] Signal matrix correction: According to the standard response function and the inverse function of the detector's pixel response function, the voltage value V of each pixel in the actual imaging signal voltage matrix V is corrected. ij , first obtain the calibration integration time t through the inverse function of the response function (i,j) , and then obtain the correction voltage through the standard response function to generate the corrected signal matrix V 校正 .

[0018] Preferably, the shape-preserving piecewise cubic interpolation method satisfies: in the subinterval [x k ,x k+1 ] using a cubic Hermite interpolation polynomial to satisfy P(x j )=y j The first-order derivative is continuous, and the second-order derivative is allowed to be discontinuous, preserving the monotonicity and local extreme value characteristics of the data.

[0019] Preferably, the array size of the detector is M×N, where M is the number of rows, N is the number of columns, and M and N are positive integers.

[0020] As a preference, in K integration times t k The pixel (i, j) under the output is V ij (t k ), whose average response is Where k = 1, 2,…, K.

[0021] As a preference, under uniform black volume integral time t, the pixel (i, j) output value V ij (t,φ) and correction value V' ij (t,φ) satisfies the mapping relation V' ij (t,φ)=f[V ij (t,φ)], φ is the radiation flux incident on the detector, V ij (t,φ) refers to the actual response output (electrical signal voltage value) of the (i,j) pixel to the uniform blackbody radiation flux φ at the integration time t, reflecting the inherent nonlinear response characteristics of the pixel; V' ij (t,φ) refers to the target output value of the (i,j)th pixel after correction.

[0022] Preferably, the correction value is:

[0023] V' ij (t,φ)=V t

[0024] Where V t is the average response corresponding to the integration time t, t∈t1,t2,…,t K .

[0025] Preferably, the paraxial approximation condition includes that the maximum field of view angle from the detector to the radiation source is less than 5°, so as to ensure that the detector surface receives uniform radiation.

[0026] As an example, the selection range of the N integration time points is [0.1×t m ,10×t n ] within the range.

[0027] Preferably, N≥9.

[0028] Preferably, the method of retaining the monotonicity and local extreme value characteristics of the data includes:

[0029] In the interval where the data is monotonically increasing or decreasing, the interpolation function remains monotonic;

[0030] Where there are local extreme points in the data, the interpolation function retains the local extreme value characteristics.

[0031] Compared with the prior art, the non-uniformity correction method for large-array infrared detectors of this application has the following beneficial effects:

[0032] The standard response function and the inverse pixel response function are generated through the conformal piecewise cubic interpolation method, which preserves the monotonicity and local extrema characteristics of the data and effectively eliminates the nonlinear response differences of the detector over a wide dynamic range. Combined with the dual-temperature calibration method, it achieves joint correction for temporal and spatial non-uniformity, with correction accuracy significantly superior to traditional methods.

[0033] Through signal matrix correction, the actual signal voltage matrix is ​​mapped to the standard response function, the response differences between pixels are eliminated, and finally a corrected signal voltage matrix is ​​generated, which significantly improves the uniformity and dynamic range of the image and reduces the non-uniformity error.

[0034] By adjusting the aperture diameter and the distance from the detector to the aperture, the paraxial approximation condition is ensured, thereby ensuring radiation uniformity and avoiding additional errors caused by uneven radiation.

[0035] The selection range of integration time points and the equal spacing conditions in logarithmic coordinates can cover the actual working scenarios of the detector; and the universal design of the detector array size M×N is suitable for infrared detectors of different sizes, without the need to adjust the algorithm framework for specific devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 This is a flowchart of the non-uniformity correction method for a large-array infrared detector provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The following is a clear and complete description of the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0039] In this document, the term "comprising" is intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.

[0040] This embodiment provides a non-uniformity correction method for a large array infrared detector. Figure 1 As shown, the non-uniformity correction method for a large-array infrared detector of this embodiment includes a reference and saturation voltage determination step, an integration time point selection step, an aperture parameter setting step, a dual-temperature calibration data acquisition step, a standard response function construction step, a pixel response inverse function solution step, and a signal matrix correction step.

[0041] Specifically, the reference and saturation voltage determination step is used to establish the voltage reference boundary of the detector response, providing the zero point and range reference for subsequent calibration. This step specifically includes:

[0042] Place the detector in a completely dark environment with no radiation input to eliminate the influence of external radiation, collect the signal voltage values ​​of all effective pixels, and calculate the average value of the signal voltage as the reference voltage, which reflects the inherent output of the detector dark current and bias circuit;

[0043] Place the detector in front of a radiation source such as a point blackbody and adjust the incident radiation intensity. When the mean signal voltage no longer increases significantly with increasing radiation intensity, it indicates that the mean signal voltage has reached its minimum value, i.e., the detector output is close to saturation. Record the mean signal voltage at this time as the saturation voltage, which corresponds to the upper threshold of the detector response.

[0044] The reference voltage and saturation voltage in this step define the effective response dynamic range of the detector, avoiding signal truncation or distortion during the calibration process.

[0045] In detail, the steps for selecting the integration time point include:

[0046] According to the actual working integration time range of the detector [t m ,t n], select N integration time points t1, t2,…, t N , and each integral time point satisfies the equal spacing condition in the logarithmic coordinate, that is, log(t n )-log(t n -1)=log(t n +1)-log(t n ).

[0047] This step adapts the nonlinear characteristics of the detector response (approximately an S-shaped curve) through logarithmic sampling, uniformly covering the response interval within a wide dynamic range, and especially accurately sampling weak signals at low integration times and saturation regions at high integration times.

[0048] Furthermore, in this step, the selection range of the N integration time points is [0.1×t m ,10×t n ] to ensure that the integration time points are evenly distributed and cover the dynamic range, thereby covering the actual working scenario of the detector, where N ≥ 9, to ensure the accuracy of the interpolation model and avoid distortion of the response curve caused by undersampling.

[0049] In detail, the steps for setting the aperture parameters include:

[0050] Adjust the integration time to the maximum integration time point t N , change the aperture diameter D and the distance L from the detector to the aperture, and make the mean value of the signal voltage close to the saturation voltage under the condition of paraxial approximation.

[0051] In this step, the detector is ensured to operate at the edge of the nonlinear region and cover the full range response characteristics by making the mean value of the signal voltage close to the saturation voltage (such as 80%-90% of the saturation voltage).

[0052] Furthermore, in this step, the paraxial approximation condition includes that the maximum field of view angle from the detector to the radiation source is less than 5°, so as to ensure that the detector surface receives uniform radiation and avoid non-uniformity test errors caused by radiation gradients.

[0053] In detail, the steps for obtaining dual-temperature calibration data include:

[0054] Keeping the aperture diameter D and the distance L from the detector to the aperture unchanged, the integration time points t1, t2, ..., t N The integration time is changed in sequence, and the signal voltage value of each effective pixel at temperature T1 (such as 300K) and T2 (such as 350K) is recorded.

[0055] This step can separate temperature drift and heterogeneity response through dual-temperature data, eliminate the influence of environmental factors on the response through temperature difference, and improve the robustness of the calibration model. It is particularly suitable for time-varying heterogeneity correction.

[0056] In detail, the steps for building a standard response function include:

[0057] Based on the integration time points t1, t2, …, t N The corresponding effective pixel signal voltage mean V t1 ,V t2 ,…,V tN , the standard response function is generated using the shape-preserving piecewise cubic interpolation method.

[0058] Furthermore, in this step, the shape-preserving piecewise cubic interpolation method satisfies: in the subinterval [x k ,x k+1 ] using a cubic Hermite interpolation polynomial to satisfy P(x j )=y j (i.e., the function values ​​at the interpolation points are consistent) and the first-order derivative is continuous, the second-order derivative is allowed to be discontinuous, and the monotonicity and local extreme value characteristics of the data are preserved. Specifically, the monotonicity and local extreme value characteristics of the data are preserved, including:

[0059] In the interval where the data is monotonically increasing or decreasing, the interpolation function remains monotonic;

[0060] Where there are local extreme points in the data, the interpolation function retains the local extreme value characteristics.

[0061] This step avoids the overshoot phenomenon of traditional interpolation by preserving the monotonicity and local extreme values ​​of the data, and accurately fits the S-shaped response curve of the detector.

[0062] In detail, the steps for solving the inverse function of pixel response include:

[0063] For the pixel (i, j) in the i-th row and j-th column, based on the integration time t ij =t1,t2,…,t N The signal voltage value under The inverse function V'=F of the response function is solved by using the shape-preserving piecewise cubic interpolation method. V (V ij ).

[0064] In this step, the actual pixel output voltage is mapped to the equivalent integration time through the inverse function, which is convenient for docking with the standard response function and realizing the mathematical conversion of non-uniformity correction.

[0065] In detail, the signal matrix correction steps include:

[0066] According to the standard response function and the inverse function of the response function of all pixels of the detector, where the array size of the detector is M×N, where M is the number of rows and N is the number of columns, and M and N are positive integers, the number of inverse functions of the response function obtained is M×N. For each pixel voltage value V in the actual imaging signal voltage matrix V ij , Perform the subsequent steps pixel by pixel;

[0067] By the inverse function of the response function V′=F V (V ij ) Get the calibration integration time t (i,j) ;

[0068] At K integration times t k The pixel (i, j) under the output is V ij (t k ), whose average response is Where k = 1, 2, ..., K;

[0069] Under uniform black volume integral time t, pixel (i, j) output value V ij (t,φ) and correction value V' ij (t,φ) satisfies the mapping relation V' ij (t,φ)=f[V ij (t,φ)], φ is the radiation flux incident on the detector, V ij (t,φ) refers to the actual response output (electrical signal voltage value) of the (i,j) pixel to the uniform blackbody radiation flux φ at the integration time t, reflecting the inherent nonlinear response characteristics of the pixel; V' ij (t,φ) refers to the target output value of the (i,j)th pixel after correction, and the correction value is V' ij (t,φ)=V t , t∈t1,t2,…,t K ; The mapping function is a nonlinear conversion function constructed by shape-preserving piecewise cubic interpolation. Its core function is to convert V ij The measured value of (t,φ) is mapped to V t , to eliminate the response differences between pixels;

[0070] Substitute the calibrated integration time into the standard response function to obtain the correction voltage and generate the corrected signal matrix V 校正 ,

[0071] This step uniformly maps the nonlinear response of each pixel onto the standard response curve to eliminate the response differences between pixels and achieve heterogeneity correction.

[0072] In summary, the non-uniformity correction method for large-array infrared detectors of this embodiment generates a standard response function and an inverse pixel response function through conformal piecewise cubic interpolation, preserving the monotonicity and local extrema characteristics of the data and effectively eliminating nonlinear response differences of the detector over a wide dynamic range. Combined with dual-temperature calibration, it achieves joint correction of temporal and spatial non-uniformity, with correction accuracy significantly superior to traditional methods. Through signal matrix correction, the actual signal voltage matrix is ​​mapped to the standard response function, eliminating response differences between pixels and ultimately generating a corrected signal voltage matrix, significantly improving image uniformity and dynamic range, and reducing non-uniformity errors. By adjusting the aperture diameter and the detector-to-aperture distance, the paraxial approximation condition is ensured, thereby ensuring radiation uniformity and avoiding additional errors introduced by radiation non-uniformity. The selection range of integration time points and the equal spacing condition in logarithmic coordinates ensure coverage of actual detector operating scenarios. Furthermore, the universal design of the detector array size M×N is adaptable to infrared detectors of different sizes, eliminating the need to adjust the algorithm framework for specific devices. Therefore, the large-array infrared detector non-uniformity correction method of this embodiment solves the difficult problem of large-scale infrared detector non-uniformity correction by combining conformal interpolation, dual-temperature calibration and logarithmic equidistant integration time point design, providing reliable technical support for high-precision infrared imaging systems.

[0073] In the embodiments provided herein, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein the communication media include any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that a computer can access. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0074] Finally, it should be noted that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent replacements for some of the technical features therein. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for correcting non-uniformity of a large array infrared detector, characterized in that: The method The following steps are involved: Determination of reference and saturation voltage: Place the detector in a completely dark environment and obtain the mean signal voltage as the reference voltage. Place the detector in front of a radiation source and adjust the incident radiation intensity so that the mean signal voltage reaches a minimum value, which is defined as the saturation voltage. Integration time point selection: According to the actual working integration time range of the detector [t m ,t n ], select N integration time points t1, t2,…, t N , and each integral time point satisfies log(t n )-log(t n -1)=log(t n +1)-log(t n ); Aperture parameter setting: Adjust the integration time to the maximum integration time point t N , change the aperture diameter D and the distance L from the detector to the aperture, and make the mean value of the signal voltage close to the saturation voltage under the condition of paraxial approximation; Dual-temperature calibration data acquisition: Keep the aperture diameter D and the distance from the detector to the aperture L unchanged, and calculate the integral time points t1, t2, ..., t N Change the integration time in sequence and record the signal voltage value of each effective pixel at temperature T1 and T2; Standard response function construction: based on the integration time points t1, t2, ..., t N The corresponding effective pixel signal voltage mean V t1 ,V t2 ,…,V tN , the standard response function is generated using the shape-preserving piecewise cubic interpolation method; Solve the inverse function of pixel response: For the pixel (i, j) in the i-th row and j-th column, based on the integration time t ij =t1,t2,…,t N The signal voltage value under The inverse function V′=F of the response function is solved by using the shape-preserving piecewise cubic interpolation method. V (V ij ); Signal matrix correction: According to the standard response function and the inverse function of the detector's pixel response function, the voltage value V of each pixel in the actual imaging signal voltage matrix V is corrected. ij , first obtain the calibration integration time t through the inverse function of the response function (i,j) , and then obtain the correction voltage through the standard response function to generate the corrected signal matrix V 校正 .

2. The method for non-uniformity correction of large array infrared detectors according to claim 1, characterized in that: The shape-preserving piecewise cubic interpolation method satisfies: in the subinterval [x k ,x k+1 ] using a cubic Hermite interpolation polynomial to satisfy P(x j )=y j The first-order derivative is continuous, and the second-order derivative is allowed to be discontinuous, preserving the monotonicity and local extreme value characteristics of the data.

3. The method for non-uniformity correction of large array infrared detectors according to claim 1, characterized in that: The size of the detector array is M×N, where M is the number of rows, N is the number of columns, and M and N are positive integers.

4. The method for correcting non-uniformity of a large array infrared detector according to claim 3, characterized in that: At K integration times t k The pixel (i, j) under the output is V ij (t k ), whose average response is Where k = 1, 2,…, K.

5. The method for non-uniformity correction of large array infrared detectors according to claim 1, characterized in that: Under uniform black volume integral time t, pixel (i, j) output value V ij (t,φ) and correction value V' ij (t,φ) satisfies the mapping relation V' ij (t,φ)=f[V ij (t,φ)], φ is the radiation flux incident on the detector, V ij (t,φ) refers to the actual response output (electrical signal voltage value) of the (i,j) pixel to the uniform blackbody radiation flux φ at the integration time t, reflecting the inherent nonlinear response characteristics of the pixel; V' ij (t,φ) refers to the target output value of the (i,j)th pixel after correction.

6. The method for correcting non-uniformity of a large array infrared detector according to claim 5, characterized in that: The correction value is: V’ ij (t,φ)=V t Where V t is the average response corresponding to the integration time t, t∈t1,t2,…,t K .

7. The method for non-uniformity correction of a large array infrared detector according to claim 1, characterized in that: The paraxial approximation condition includes that the maximum field angle from the detector to the radiation source is less than 5°, so as to ensure that the detector surface receives uniform radiation.

8. The method for correcting non-uniformity of a large array infrared detector according to claim 1, characterized in that: The selection range of N integration time points is [0.1×t m ,10×t n ] within the range.

9. The method for correcting non-uniformity of a large array infrared detector according to claim 8, characterized in that: N≥9。 10. The method for correcting non-uniformity of a large array infrared detector according to claim 2, characterized in that: The monotonicity and local extreme value characteristics of the preserved data include: In the interval where the data is monotonically increasing or decreasing, the interpolation function remains monotonic; Where there are local extreme points in the data, the interpolation function retains the local extreme value characteristics.

Images