An infrared focal plane detector flash element test method based on dual thresholds

Abstract

The present invention discloses a method for testing infrared focal plane detector flash elements based on a dual threshold value, which relates to the technical field of infrared focal plane detector performance testing. First, the temperature of a surface source black body is set, and the output voltage of the infrared focal plane detector at the temperature is measured; the F-frame response data of the infrared focal plane detector is continuously collected; the blind pixels in the focal plane are screened and eliminated; among the remaining pixels in the focal plane, the pixels whose pixel noise is greater than a specified threshold σ1 are marked as time noise flash elements, and the pixels whose pixel volatility parameter is greater than a specified threshold σ2 are marked as volatility flash elements, and their positions are recorded; the union of the time noise flash elements and the volatility flash elements is taken as the focal plane flash element. The present invention simultaneously focuses on the temporal noise of the pixels in the focal plane and the overall volatility of the pixel time domain response, avoids the misdetection and missed detection of flash elements with different characteristics, and realizes the effective detection of flash elements that affect the working accuracy and stability of the infrared focal plane detector.

Description

Technical Field

[0001] The present invention relates to the technical field of infrared focal plane detector performance testing, and more particularly to a flash element testing method of an infrared focal plane detector based on a double threshold value. Background Art

[0002] Infrared detection has a wide range of applications in both military and civilian fields, such as infrared night vision, earth observation, engineering inspection, and meteorological monitoring. This drives increasingly stringent requirements for the detection quality and accuracy of infrared focal plane (FPA) detectors. Due to the influence or limitations of materials, processes, and operating environments, some pixels exhibit excessively high or low response during imaging, known as blind pixels. Compensation for these blind pixel responses is necessary in practical FPA applications. With the advancement of FPA technology, the relevant testing methods, evaluation indicators, and correction and compensation for blind pixel performance have become increasingly sophisticated. These principles are primarily based on the definitions in the national standard GB / T17444-2013, "Infrared Focal Plane Array Parameter Test Method."

[0003] A flicker pixel is defined as a pixel that behaves as a blind pixel less than 100% of the time during the detector's operating cycle. As an inherent characteristic of infrared detectors, flicker pixels significantly impact their target detection and imaging performance. Over a certain timeframe, flicker pixel output levels fluctuate significantly, sometimes returning to normal. Due to their time-domain fluctuations, flicker pixels are a difficult factor to detect and characterize in detector performance analysis. Unlike blind pixels, their response information still holds value. Due to the irregular nature of their fluctuations, the mean response of a flicker pixel may be similar to that of a normal pixel. Treating them as blind pixels could result in false or missed target detections, necessitating the design of new testing methods to extract flicker pixel information.

[0004] Flash pixels appear less frequently than blind pixels and are difficult to quantitatively evaluate based on their definition. Therefore, testing for flash pixels is difficult, and research on flash pixels is not very common, with only a small number of researchers conducting research. Existing flash pixel detection methods typically directly compare the pixel output signal value with the absolute value of the single-frame output signal average and a specified threshold, while ignoring the temporal fluctuations of the flash pixels. This can easily lead to low accuracy in flash pixel detection results. Therefore, providing a flash pixel testing method for infrared focal plane detectors that takes into account temporal noise and pixel temporal response fluctuations is an urgent problem that those skilled in the art need to solve. Summary of the Invention

[0005] In view of this, the present invention provides a flash pixel testing method for an infrared focal plane detector based on a dual threshold, which simultaneously focuses on the temporal noise of the pixel in the focal plane and the overall volatility of the pixel's time domain response. Through this method, the position of the flash pixel can be effectively detected.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A double-threshold-based method for testing flash elements in an infrared focal plane detector comprises the following steps:

[0008] Step 1: setting the temperature T0 of the surface source blackbody according to the response wavelength of the infrared focal plane detector, and measuring the output voltage of the infrared focal plane detector at temperature T0;

[0009] Step 2, continuously collecting F-frame response data of the infrared focal plane detector;

[0010] Step 3: Screen and remove blind pixels in the focal plane;

[0011] Step 4: Mark the pixel noise V in the remaining pixels on the focal plane N Pixels larger than the specified threshold σ1 are temporal noise flash pixels, and their positions are recorded;

[0012] Step 5: Mark the pixel volatility parameter V in the remaining pixels on the focal plane F Pixels with a value greater than the specified threshold σ2 are considered as fluctuating flash pixels, and their positions are recorded;

[0013] Step 6: Take the union of all time noise flash elements and volatility flash elements, which is the focal plane flash element.

[0014] Optionally, in step 3, the method for screening and eliminating blind pixels in the focal plane is:

[0015] The output signal voltage, response rate and noise of the infrared focal plane detector are calculated, blind pixels are screened according to the preset blind pixel criterion, the number of dead pixels d and the number of overheated pixels h are counted, and the blind pixels in the focal plane are eliminated.

[0016] Optionally, in step 4, the pixel noise V N The calculation formula is:

[0017]

[0018] Among them, F is the number of response data frames collected, V S [(i,j),T0,f] is the response data of the pixel (i,j) at the temperature T0 in the fth frame. is the mean of all F-frame response data of pixel (i, j) at temperature T0.

[0019] Optionally, the number of response data frames F ≥ 10000.

[0020] Optionally, in step 4, the calculation method of the threshold σ1 is specified as follows:

[0021]

[0022] Where A is a constant; is the focal plane average noise voltage, which is calculated as:

[0023]

[0024] Among them, M is the total number of pixel columns, N is the total number of pixel rows, d is the number of dead pixels in the focal plane, and h is the number of overheated pixels in the focal plane.

[0025] Optionally, in step 5, the pixel volatility parameter V F The calculation method is:

[0026] Step 5.1, divide the continuous F frame response data into n segments, each segment contains m frames of data;

[0027] Step 5.2: For each pixel (i, j), calculate the response mean of n segments of data Differences of adjacent means, Get the new difference sequence {ΔV1(i,j),ΔV2(i,j),...,ΔV n-1 (i,j)};

[0028] Step 5.3, for the obtained differential sequence {ΔV1(i,j),ΔV2(i,j),...,ΔV n-1 (i, j)} calculate the standard deviation to get the pixel volatility parameter V F (i,j).

[0029] Optionally, in step 5.3, the pixel volatility parameter V F (i,j) is:

[0030]

[0031] Where, ΔV l (i, j) is the difference between the data of segment (l+1) and segment l, ΔV l (i,j) difference series mean.

[0032] As can be seen from the above technical solution, the present invention provides a double-threshold-based method for testing flash elements in an infrared focal plane detector, which has the following beneficial effects compared with the prior art:

[0033] Compared with traditional testing methods, this method simultaneously considers the flicker characteristics of both pixel temporal noise and the overall volatility of the pixel's temporal response. This avoids false detection and missed detection of flicker pixels with different characteristics, effectively detecting flicker pixels that affect the accuracy and stability of infrared focal plane detectors. This method has been proven to be applicable to a variety of infrared focal plane detectors, providing reliable and effective flicker detection results. It provides a testing method for studying the temporal stability of infrared focal plane detectors, offers valuable insights for reducing the number of flicker pixels in infrared focal plane devices, and is widely applicable to the performance testing of infrared detectors. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention 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 merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0035] Figure 1 This is a schematic diagram of the structure of the infrared focal plane detector flash element test system based on dual thresholds in the present invention;

[0036] Figure 2 This is a flow chart of the flash element testing method of an infrared focal plane detector based on dual thresholds of the present invention;

[0037] Figure 3 This is a grayscale image of the original test result obtained by the flash element test system in a specific embodiment;

[0038] FIG4( a ) is a schematic diagram of the output signal voltage of a normal element for 50,000 frames in a specific embodiment;

[0039] FIG4( b ) is a schematic diagram of the output signal voltage of the first flash element at 50,000 frames in a specific embodiment;

[0040] FIG4( c ) is a schematic diagram of the output signal voltage of the second flash element at 50,000 frames in a specific embodiment;

[0041] FIG4( d ) is a schematic diagram of the output signal voltage of the third flash element at 50,000 frames in a specific embodiment;

[0042] Figure 5 This is a distribution diagram of flash elements measured in a specific embodiment;

[0043] Among them, 1 is a surface source blackbody, 2 is an infrared focal plane detector, 3 is a detector driver board, 4 is a power supply bias circuit, 5 is a clock driver circuit, 6 is an analog data acquisition card, and 7 is a computer server. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0045] The embodiment of the present invention discloses a double-threshold infrared focal plane detector flash element testing method, see Figure 2 , including the following steps:

[0046] Step 1: setting the temperature T0 of the surface source blackbody 1 according to the response wavelength of the infrared focal plane detector 2, and measuring the output voltage of the infrared focal plane detector 2 at the temperature T0;

[0047] Step 2, continuously collecting F frame response data of the infrared focal plane detector 2;

[0048] Step 3: Calculate the output signal voltage, response rate, and noise of the infrared focal plane detector 2, filter blind pixels according to the preset blind pixel criterion, count the number of dead pixels d and the number of overheated pixels h, and eliminate blind pixels in the focal plane;

[0049] Step 4: Mark the pixel noise V in the remaining pixels on the focal plane N Pixels larger than the specified threshold σ1 are temporal noise flash pixels, and their positions are recorded.

[0050] Among them, pixel noise V N The calculation formula is:

[0051]

[0052] Where, F is the number of response data frames collected, F≥10000, V S [(i, j), T0, f] is the response data of the pixel (i, j) at the temperature T0 in the fth frame. is the mean of all F-frame response data of pixel (i, j) at temperature T0.

[0053] The calculation method of the prescribed threshold σ1 is: Where A is a constant, taking the value of 3, 5, 6, 8 or 10; is the average noise voltage of the focal plane, which represents the noise of all remaining pixels after removing the blind pixels. N The average value is calculated as:

[0054]

[0055] Where M is the total number of columns of pixels, N is the total number of rows of pixels, d is the number of dead pixels in the focal plane, and h is the number of overheated pixels in the focal plane.

[0056] Step 5: Mark the pixel volatility parameter V in the remaining pixels on the focal plane F Pixels with a value greater than the specified threshold σ2 are considered as fluctuating flash pixels, and their positions are recorded.

[0057] Among them, the pixel volatility parameter V F The calculation method is:

[0058] Step 5.1: Divide the continuous F-frame response data into n segments, each containing m frames of data. The values of m and n can be set according to actual needs. m represents the frequency resolution of the flash element time fluctuation analysis. The smaller m is, the higher the fluctuation frequency is. n is the corresponding number of data segments.

[0059] Step 5.2: For each pixel (i, j), calculate the mean response value of n segments of data. Differences of adjacent means, Get the new difference sequence {ΔV1(i,j),ΔV2(i,j),...,ΔV n-1 (i,j)};

[0060] Step 5.3, for the obtained differential sequence {ΔV1(i,j),ΔV2(i,j),...,ΔV n-1 (i, j)} calculate the standard deviation to get the pixel volatility parameter V F (i,j):

[0061]

[0062] Where ΔV l (i, j) is the difference between the data of segment (l+1) and segment l, ΔV l (i,j) difference series mean.

[0063] The selection of the threshold value σ2 used for judgment is related to the actual test data. It is generally believed that the pixel volatility parameter V F The distribution of (i,j) approximately satisfies the normal distribution N(μ,σ 2 ), where μ is the mathematical expectation of the distribution, σ 2 is the variance, so combined with the actual response data V S [(i, j), T0, f] to design the threshold σ2, refer to σ2 = kσ, to achieve V F It is used as a benchmark for flash element testing of different accuracies.

[0064] Step 6: The final flash detection result is obtained through double threshold synthesis. The union of all temporal noise flashes and volatility flashes is the focal plane flash.

[0065] The following describes the specific implementation of the present invention in further detail, taking the flash element test of a 320×256 medium-wave infrared focal plane detector as an example:

[0066] 1. Building a dual-threshold infrared focal plane detector flash element test system

[0067] Build a system for testing flash elements of 320×256 medium-wave infrared focal plane detectors, such as Figure 1 As shown, it mainly includes a surface source blackbody 1, an infrared focal plane detector to be tested 2, a detector driver board 3, a power bias circuit 4, a clock driver circuit 5, an analog data acquisition card 6, a computer server 7 and Figure 1 The optical test platform, power supply, etc. are not shown.

[0068] The infrared focal plane detector 2 to be tested is encapsulated in a liquid nitrogen dewar and fixed on an optical test platform. The detector driver board 3 is fixed on the platform and connected to the infrared focal plane detector 2 dewar through a test interface.

[0069] At the start of the test, the power bias circuit 4 and clock drive circuit 5, under the control of the computer server 7, send bias and clock signals to the driver board. The test interface then provides a normal operating environment for the infrared focal plane detector 2. After the detector driver board is turned on, data is collected. After sample and hold, the collected data is transmitted back to the analog data acquisition card 6 via a data line. A / D conversion is performed in subsequent circuits, and the resulting digital signal is transmitted to the computer server 7. System control and data processing software converts the signal data into a response result, which is then fed back as a grayscale image in real time and stored.

[0070] 2. Testing process of infrared focal plane detector flash element

[0071] After configuring the flash test system, perform the flash test according to the following steps:

[0072] Step (1): Since the response wavelength of the infrared focal plane detector 2 to be tested is medium-wave infrared, a suitable black body temperature T0=20°C is set, and the output voltage of the infrared focal plane detector 2 at the temperature of each pixel can be obtained by the test system. Figure 3 The grayscale image of the response obtained from the test is shown;

[0073] Step (2): continuously collect F = 50,000 frames of data, and record the response voltage of pixel (i, j) at temperature T0 in the f-th frame of data as V S [(i,j),T0,f];

[0074] Step (3), according to the GB / T-17444-2013 test method, the output signal voltage, response rate, and noise of the infrared focal plane detector 2 are calculated, blind pixels are screened according to the specified blind pixel criterion, and the number of dead pixels d and the number of overheated pixels h are counted;

[0075] Step (4): Select the response voltage at temperature T0 as the standard for flash pixel test. After removing the blind pixels in the focal plane, calculate the noise V of the remaining pixels in each focal plane at the blackbody temperature T0. N , focal plane average noise voltage Calculate according to formula (1) and formula (2) respectively:

[0076]

[0077]

[0078] In the formula is the average value of 50,000-frame response of pixel (i, j) at T0 = 293 K, M = 320 is the total number of pixel columns, N = 256 is the total number of pixel rows, d and h are the number of dead pixels and overheated pixels of the filtered infrared focal plane detector.

[0079] Step (5): Since the response of the flash pixel has a time-varying fluctuation characteristic, its detection needs to be based on long-term test results. According to the relevant definition of the flash pixel, its temporal noise will be significantly increased compared to the normal pixel, and the temporal response of the flash pixel will also show a fluctuating characteristic. Based on this, the detection of the flash pixel is achieved through a dual threshold. The two thresholds focus on the temporal noise of the pixel and the overall volatility of the temporal response of the pixel. Specifically, the temporal noise flash pixel is tested by the threshold σ1, and the threshold σ1 is calculated according to formula (3):

[0080]

[0081] Where A is a custom constant that can be set according to the actual need to test the instability of the flash element. Since the time stability of the focal plane to be tested and the test system as a whole is good, A=3 is taken. For each pixel (i, j), if The pixel (i, j) is marked as a temporal noise flash pixel, and 25 temporal noise flash pixels are screened out in this way.

[0082] Step (6) tests the fluctuation flash element by threshold σ2. The specific test steps are as follows: divide the response of the focal plane continuous F frame data into n segments, each segment contains m frames of data. In this embodiment, m=8, corresponding to n=6250, while ensuring the local fluctuation characteristics, the test speed is increased as much as possible. For each pixel (i, j), the response mean of its 6250 segments of data is calculated. Difference the adjacent means to obtain a new difference sequence Then the obtained differential result sequence {ΔV1(i,j),ΔV2(i,j),...,ΔV 6249 (i, j)} calculate the standard deviation to get the pixel volatility parameter V F (i, j), calculated according to formula (4):

[0083]

[0084] Select a suitable threshold σ2, for pixel (i, j), when the pixel volatility parameter V F When (i, j) is greater than σ2, the pixel is determined to be a fluctuating pixel.

[0085] The selection of threshold σ2 is related to the actual test data. Due to the randomness of the pixel fluctuation, it is generally believed that the pixel fluctuation parameter V F The distribution of (i,j) approximately satisfies the normal distribution N(μ,σ 2 ), where μ is the mathematical expectation of the distribution, σ 2 is the variance, so combined with the actual response data V S [(i,j),T0,f] is used to design the threshold σ2, and any normal pixel (i0,j0) is selected. Gaussian fitting is performed on it to extract the normally distributed σ(i0,j0) information. In order to ensure the consistency of the detection standard, σ2 takes a fixed value near 3σ(i0,j0) as the standard for all pixels to detect fluctuating flashes. The screening standard in this embodiment is σ2=0.5mV, and 19 fluctuating flashes are detected.

[0086] The final flash detection result is obtained by combining the double thresholds. The union of the time noise flash and the volatility flash is taken to get the focal plane flash test result, which is a total of 29 flashes. Figure 4(a)-Figure 4(d) The time domain comparison of the output signal voltage of normal cells and some flash cells over 50,000 frames is given. Figure 5 This is the distribution diagram of the 29 flash elements obtained from the test. Figure 5 The middle black dot corresponds to the flash element position.

[0087] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0088] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A double-threshold infrared focal plane detector flash element testing method, characterized in that: The following steps are involved: Step 1: Set the temperature of the surface source blackbody according to the response wavelength of the infrared focal plane detector , and measured the infrared focal plane detector at temperature Output voltage under Step 2, continuously collecting F-frame response data of the infrared focal plane detector; Step 3: Screen and remove blind pixels in the focal plane; Step 4: Mark the pixel noise in the remaining pixels in the focal plane Greater than the specified threshold The pixel is a temporal noise flash pixel, and the position is recorded. The calculation formula is: ; Where F is the number of response data frames collected, For pixels exist The response data of the fth frame at temperature, For pixels exist The mean value of all F-frame response data at temperature; Step 5: Mark the pixel volatility parameters in the remaining pixels on the focal plane Greater than the specified threshold The pixel is a fluctuation flash pixel, and the position and pixel fluctuation parameters are recorded. The calculation method is: Step 5.1, divide the continuous F frame response data into n segments, each segment contains m frames of data; Step 5.2: For each pixel , respectively find the response mean of n segments of data , for the adjacent mean differences, , and obtain a new difference sequence ; Step 5.3: Get the difference sequence Calculate the standard deviation to obtain the pixel volatility parameter ; Step 6: Take the union of all time noise flash elements and volatility flash elements, which is the focal plane flash element.

2. The method for testing flash elements of an infrared focal plane detector based on a dual threshold according to claim 1, characterized in that: In step 3, the method for screening and eliminating blind pixels in the focal plane is: The output signal voltage, response rate and noise of the infrared focal plane detector are calculated, blind pixels are screened according to the preset blind pixel criterion, the number of dead pixels d and the number of overheated pixels h are counted, and the blind pixels in the focal plane are eliminated.

3. The method for testing flash elements of an infrared focal plane detector based on a dual threshold according to claim 1, characterized in that: Number of response data frames .

4. The method for testing flash elements of an infrared focal plane detector based on a dual threshold according to claim 1, characterized in that: In step 4, the threshold value is set The calculation method is: ; Where A is a constant; is the focal plane average noise voltage, which is calculated as: ; Among them, M is the total number of pixel columns, N is the total number of pixel rows, d is the number of dead pixels in the focal plane, and h is the number of overheated pixels in the focal plane.

5. The method for testing flash elements of an infrared focal plane detector based on dual thresholds according to claim 1, characterized in that: In step 5.3, the pixel volatility parameter for: ； in, For the first for the first ( l +1) Paragraph and l The difference result of the segment data, for Differenced series mean.

Images