Infrared detector focal plane pixel evaluation and selection method and device based on z fraction

Abstract

The invention discloses an infrared detector focal plane pixel evaluation and selection method and device based on a z score. The infrared detector focal plane pixel evaluation and selection method comprises the steps of obtaining to-be-evaluated data of each pixel in an infrared focal plane array device; preprocessing the to-be-evaluated data to filter bad pixels to obtain performance index data of each pixel after filtering; calculating an average value and a standard deviation of each performance index of all the filtered pixels; calculating a z score of the performance index corresponding to each pixel, and summarizing and adding the z scores of the performance indexes corresponding to each pixel to obtain a final z score result of the comprehensive performance index of each pixel; and selecting the pixel with the highest z score in each row of the infrared focal plane array device according to the final z score result of each pixel, and registering the selected pixel with the highest z score as the optimal pixel, so that all performance indexes are integrated, the comprehensively evaluated and registered optimal pixel is obtained, each index is improved to a certain extent, and the data quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of infrared detector signal processing technology, and in particular to a method and device for evaluating and selecting focal plane pixels of infrared detectors based on z-scores, which is particularly suitable for long linear array detectors with multiple columns of redundant pixels in the field of satellite remote sensing. Background Art

[0002] Long-line array detectors refer to large-scale long-line array focal plane detector scanning radiation imagers, which use push-scanning to detect two-dimensional optical signals, such as infrared images, visible light images, etc. Large-scale long-line array focal plane detector scanning radiation imagers are currently widely used in the field of satellite remote sensing. Due to the influence of manufacturing materials, processes and other factors (such as material non-uniformity, mask errors, defects, etc.), the performance of each pixel in the infrared focal plane array device is different, and there are even blind pixels that cannot be used normally. The number and distribution of pixels and blind pixels with poor performance have a great impact on the signal-to-noise ratio and image quality of infrared images. If there are too many blind pixels or their distribution is too concentrated, a large number of or too concentrated bad lines will appear on the infrared image of the long-line array detector.

[0003] In order to improve the imaging quality of infrared detectors and avoid the influence of poor-performance pixels and interference of blind pixels in infrared detector arrays, a variety of technical solutions have been proposed. In addition to improving the manufacturing process of infrared detectors, a common and effective method is to optimize the redundant pixels.

[0004] For example, the "multi-column selection" scheme adopted by payloads such as the United States ABI, Japan AHI, South Korea AMI and Europe FCI, that is, for multi-line focal plane detectors, under the condition of meeting the observation requirements, only the pixel data with the best output response characteristics in each row is selected for image formation. This avoids the influence of blind pixels and greatly improves the accuracy and quality of the data.

[0005] The evaluation indicators related to the radiation dimension quality of infrared remote sensing images of scanning radiation imagers are mainly sensitivity and non-uniformity. Among them, sensitivity characterizes the temperature resolution ability of each pixel in the focal plane, and its value is mainly limited by the pixel noise intensity. The sensitivity indicators generally include noise voltage noise, noise equivalent temperature difference NEΔT, and ratio detection rate D*. The greater the signal noise obtained by the pixel with poor sensitivity index, the worse the signal-to-noise ratio; non-uniformity refers to the difference in physical quantities obtained by each pixel when each pixel in the focal plane images the same radiation target, which is reflected as spatial noise between pixels in remote sensing images. The uniformity indicators generally include response rate R, signal voltage change ΔV, dark background response DNbg, etc. The worse the uniformity index of the pixel, the greater the spatial noise of the obtained signal, and it will limit the amplification factor of the amplifier circuit in the detector, thereby reducing the data accuracy and sensitivity. Therefore, how to comprehensively consider various indicators and select the optimal pixel from each row is an issue that needs to be studied urgently.

[0006] Li Wenli. Research on key technologies of radiation imaging of space-borne infrared long-line array detectors [D]. Shanghai Institute of Technical Physics, Chinese Academy of Sciences, 2022. In this paper, a method for evaluating the optimal pixel that takes into account both sensitivity and uniformity is proposed. The input data of the algorithm are the instrument's on-orbit blackbody observation data, cold space observation data, and uniform area observation data. The algorithm is mainly divided into two steps, namely preliminary selection and fine selection. The defect is that when the instrument is in orbit, the pixel evaluation is performed after obtaining the on-orbit observation data. When the instrument is not yet in orbit or the infrared detector module has not been assembled into the instrument, the existing technology cannot evaluate the performance of each pixel and select the best one; the existing technical solution uses an iterative cycle method to select the optimal pixel, which is relatively complicated. Summary of the invention

[0007] The purpose of the present invention is to provide a method and device for evaluating and selecting focal plane pixels of infrared detectors based on z-scores, setting an evaluation scheme based on different indicators, evaluating different pixels, and selecting the best pixel in each row.

[0008] The present invention provides a method for evaluating and selecting infrared detector focal plane pixels based on z-scores, comprising:

[0009] Acquire the data to be evaluated of each pixel in the infrared focal plane array device;

[0010] Preprocessing the data to be evaluated to filter bad pixels, and obtaining performance index data of each pixel after filtering;

[0011] Calculate the mean and standard deviation of each performance indicator for all pixels after filtering;

[0012] The z score of the performance index corresponding to each pixel is calculated according to the mean value and standard deviation of the performance index corresponding to each pixel, and the z scores of each performance index corresponding to each pixel are summarized and added to obtain the final z score result of the comprehensive performance index of each pixel;

[0013] According to the final z-score result of each pixel, the pixel with the highest z-score in each row of the infrared focal plane array device is selected and aligned as the optimal pixel.

[0014] Preferably, the data to be evaluated include the noise voltage noise of each pixel, the noise equivalent temperature difference NEΔT, the detection ratio D*, the response rate R, the signal voltage change ΔV, the dark background response DNbg and performance indicators related to different test conditions.

[0015] Preferably, the step of obtaining the data to be evaluated for each pixel in the infrared focal plane array device comprises:

[0016] Based on the focal plane array device, observe the black body and record the current temperature T1, obtain the effective response value of each pixel, and calculate the noise root mean square noise1 at the black body temperature;

[0017] Based on the focal plane array device, observe the black body and record the current temperature T2, obtain the effective response value of each pixel, and calculate the noise root mean square noise2 at the black body temperature;

[0018] Calculate the noise equivalent temperature difference NEΔT:

[0019]

[0020] Among them, DN1 and DN2 are both valid response values.

[0021] Preferably, the preprocessing of the data to be evaluated to filter bad pixels to obtain performance indicator data of each pixel after filtering includes:

[0022] Acquire the data to be evaluated and calculate the performance index of each pixel;

[0023] Identify target pixels that meet any of the following index conditions as bad pixels and filter out the bad pixels;

[0024] Wherein, the target pixels include:

[0025] When the performance indicator of the current pixel is the response rate R, the response rate R is less than 0.5 times the average response rate of the focal plane pixel;

[0026] When the performance indicator of the current pixel is the response rate R, the pixel whose response rate R is greater than 1.5 times the average response rate of the focal plane pixel;

[0027] When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is less than 0.5 times the average response value of the focal plane pixel;

[0028] When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is 1.5 times greater than the average response value of the focal plane pixel;

[0029] When the performance indicator of the current pixel is the noise voltage noise, the pixel whose noise voltage noise is greater than twice the average noise voltage of the focal plane pixel;

[0030] When the performance index of the current pixel is the noise equivalent temperature difference NEΔT, the pixel whose noise equivalent temperature difference NEΔT is greater than the preset threshold;

[0031] The bad pixels include blind pixels, overheated pixels, flickering pixels and related data that affect the performance and image quality of the image sensor.

[0032] Preferably, the step of calculating the z score of the performance indicator corresponding to each pixel according to the average value and standard deviation of the performance indicator corresponding to each pixel includes:

[0033] The z-score calculation method for each performance index index of each pixel i is as follows:

[0034] For positive indicators, the z-score is:

[0035]

[0036] For negative indicators, the z-score is:

[0037]

[0038] For the uniformity index, the z-score is

[0039]

[0040] Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and O. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, index i The data of the pixel at the observation point or data point.

[0041] Preferably, the z-score of each performance indicator corresponding to each pixel is aggregated and added to obtain the final z-score result of the comprehensive performance indicator of each pixel, including:

[0042] The final z-score for the overall performance metric for each pixel is calculated as follows:

[0043]

[0044] Among them, k(index) is the weight coefficient of each indicator.

[0045] The present invention provides a device for evaluating and selecting infrared detector focal plane pixels based on z-scores, comprising:

[0046] A data acquisition module, used for acquiring the data to be evaluated of each pixel in the infrared focal plane array device;

[0047] A preprocessing module, used for preprocessing the data to be evaluated to filter out bad pixels, and obtaining performance index data of each pixel after filtering;

[0048] The first calculation module is used to calculate the average value and standard deviation of each performance index of all pixels after filtering;

[0049] The second calculation module is used to calculate the z score of the performance index corresponding to each pixel according to the average value and standard deviation of the performance index corresponding to each pixel, and summarize and add the z scores of each performance index corresponding to each pixel to obtain the final z score result of the comprehensive performance index of each pixel;

[0050] A registration module is selected to select the pixel with the highest z score in each row of the infrared focal plane array device according to the final z score result of each pixel, and register it as the optimal pixel.

[0051] Preferably, the second calculation module includes:

[0052] The z-score calculation method for each performance index index of each pixel i is as follows:

[0053] For positive indicators, the z-score is:

[0054]

[0055] For negative indicators, the z-score is:

[0056]

[0057] For the uniformity index, the z-score is

[0058]

[0059] Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and O. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, index i The data of the pixel at the observation point or data point.

[0060] The present invention also provides a computer device, comprising:

[0061] A memory, the memory being used to store a processing program;

[0062] A processor, wherein when executing the processing program, the processor implements the infrared detector focal plane pixel evaluation and selection method based on z-score as described in an embodiment of the present invention.

[0063] The present invention also provides an electronic device, comprising:

[0064] A memory, the memory being used to store a processing program;

[0065] A processor, wherein when executing the processing program, the processor implements a method for evaluating and selecting infrared detector focal plane pixels based on z-scores as described in an embodiment of the present invention.

[0066] With respect to the prior art, the present invention has the following beneficial effects:

[0067] The present invention proposes a method for evaluating and selecting focal plane pixels of infrared detectors based on z-scores, that is, a pixel evaluation and optimal pixel selection scheme based on z-scores. After obtaining the performance index of each pixel, the z-score of each index and the z-score of each pixel are calculated and the optimal pixel is selected. After the pixel evaluation and selection, all indicators of the detector are improved to varying degrees. The algorithm has low complexity and is easy to operate on orbit. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 This is a schematic diagram of a pixel scale of 256 rows and 4 columns in the focal plane of a long linear array infrared detector in one embodiment of the present invention;

[0069] Figure 2 A schematic diagram of a process for selecting a detector with the best performance in one embodiment of the present invention;

[0070] Figure 3 This is an example diagram of the steps of evaluating and selecting infrared detector focal plane pixels based on z-scores in one embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.

[0072] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0073] It should be noted that the concepts such as "once" and "second" mentioned in the present application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0074] It should be noted that the modifications of "one" and "plurality" mentioned in the disclosure of this application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0075] Embodiment 1

[0076] Due to the influence of manufacturing materials, processes and other factors such as material non-uniformity, mask errors, defects, etc., the performance of each pixel in the infrared focal plane array device is different, and there are even blind pixels that cannot be used normally. The number and distribution of pixels with poor performance and blind pixels have a great impact on the signal-to-noise ratio and image quality of infrared images. If there are too many blind pixels or they are too concentrated, a large number of or too concentrated bad pixels will appear on the infrared image.

[0077] In order to improve the imaging quality of infrared detectors and avoid the influence of poor-performance pixels and interference of blind pixels in the infrared detector array, some long-array infrared focal plane detectors adopt a multi-column redundant structure design to ensure the availability of the infrared remote sensing data they acquire. Figure 1 As shown, the focal plane of the long-line array infrared detector adopts a pixel scale of 256 rows and 4 columns. In actual use, only the data of a certain optimal pixel in each row is selected and used, thereby improving data quality.

[0078] In addition to selecting the best pixels for image formation when the instrument is in orbit, the performance of the detector needs to be evaluated on the ground after the detector is prepared before the instrument is launched to determine whether the detector meets the launch requirements, or to select the detector with the best performance from the multiple prepared detectors. At this stage, it is necessary to select the best pixels for the detector and then evaluate the best pixels to evaluate whether the detector meets the technical indicators, see Figure 2 shown.

[0079] There are many evaluation indicators for pixels, which can be generally divided into two categories: sensitivity and uniformity. Sensitivity indicators can measure the detection ability and accuracy of pixels. The indicators generally include noise voltage, noise equivalent temperature difference NEΔT, and detection rate D*. Uniformity indicators can measure the uniformity and consistency of different pixels. The uniformity indicators generally include response rate R, signal voltage change ΔV, dark background response DN, etc. bg Etc. The problem to be solved by the embodiment of the present invention is, after obtaining these indicators, how to set up an evaluation scheme by integrating different indicators, evaluate different pixels, and select the optimal pixel in each row. The present invention proposes a pixel evaluation and optimal pixel selection scheme based on z-scores. After obtaining the performance indicators of each pixel, the z-score of each indicator and the z-score of each pixel are calculated and the optimal pixel is selected. After pixel evaluation and selection, all indicators of the detector are improved to varying degrees. The algorithm has low complexity and is easy to run on orbit.

[0080] like Figure 3 As shown, the present invention provides a method for evaluating and selecting infrared detector focal plane pixels based on z-scores, which is suitable for improving performance in on-orbit operation scenarios, including:

[0081] S1: Obtain the data to be evaluated for each pixel in the infrared focal plane array device, perform performance tests on the infrared focal plane array device, and obtain the performance indicators of each pixel. Depending on the test plan and requirements, the performance indicators obtained are different. Here, the data to be evaluated include the noise voltage noise, noise equivalent temperature difference NEΔT, detection ratio D*, response rate R, signal voltage change ΔV, dark background response DNbg of each pixel, and performance indicators related to different test conditions;

[0082] 1. The detailed description of obtaining the data to be evaluated for each pixel in the infrared focal plane array device is as follows:

[0083] 1.1 Noise voltage

[0084] The noise voltage can measure the sensitivity and stability of a pixel. Generally speaking, the smaller the noise voltage, the stronger the detection ability of the pixel.

[0085] The test and calculation method is to use a focal plane array device to observe a black body with a temperature of T, obtain multiple effective response values ​​of each pixel, generally more than 1000 times, and calculate the root mean square noise (I, T) at the black body temperature:

[0086]

[0087] Where DN f (i, T) is the effective response value of the f-th observation of pixel i, and the unit is generally mV. is the average value of all f observations. The noise voltage noise is what we are looking for.

[0088] 1.2 Noise equivalent temperature difference NEΔT

[0089] For the infrared band of the radiation imager, the noise equivalent temperature difference NEΔT is often used to characterize the system sensitivity, which is defined as the change in the temperature of the black body radiation source at a specified temperature when the difference in the system output signal is equal to the noise root mean square of the output signal. It can be intuitively understood as how much temperature measurement accuracy error is caused by the noise voltage noise. The noise equivalent temperature difference NEΔT can measure the sensitivity of a pixel. Generally speaking, the smaller the noise equivalent temperature difference NEΔT, the stronger the detection ability of the pixel.

[0090] For a certain pixel i of the infrared detector in the infrared imaging system, there are many ways to calculate its noise equivalent temperature difference, the commonly used ones are as follows:

[0091] (1) First, calculate the noise voltage noise according to the method in Section 1.1;

[0092] (2) Calculate the noise equivalent radiance NEΔL(i,T)

[0093]

[0094] Where L(T) refers to the average spectral radiance of the black body received by the system, and the calculation formula is as follows:

[0095]

[0096] Where RSR i (λ) is the spectral response function of pixel i, L BB (λ, T) is the blackbody radiance.

[0097] (3) Calculate the average spectral differential radiance at temperature T

[0098]

[0099] (4) Finally, the noise equivalent temperature difference NEΔT is obtained, the unit is generally mK

[0100]

[0101] However, in actual measurement and engineering applications, the above calculation method is very cumbersome and difficult to apply. Therefore, the following approximate method is often used to measure the noise equivalent temperature difference NEΔT of a certain pixel.

[0102] Based on the focal plane array device, observe the black body and record the current temperature T1, obtain the effective response value of each pixel, which should generally be greater than 1000 times, and calculate the noise root mean square noise1 at the black body temperature;

[0103] Based on the focal plane array device, observe the black body and record the current temperature T2. The difference between T1 and T2 should be small, about 5K. Obtain the effective response value of each pixel, which should generally be greater than 1000 times. Calculate the noise root mean square noise2 at the black body temperature.

[0104] Calculate the noise equivalent temperature difference NEΔT:

[0105]

[0106] Among them, DN1 and DN2 are both valid response values.

[0107] The noise equivalent temperature difference NEΔT is what we are looking for. It should be noted that the noise equivalent temperature difference NEΔT is generally different at different temperatures, so the noise equivalent temperature difference NEΔT is a function of temperature T.

[0108] 1.3 Detection ratio D*

[0109] The detection rate D* can measure the detection capability of a pixel. Generally speaking, the larger the D*, the better.

[0110] The specific detection rate D* is calculated from the noise equivalent power NEP and the detection rate D. The measurement and calculation steps of the specific detection rate D* are as follows:

[0111] (1) First calculate the noise equivalent power NEP.

[0112] If the output voltage generated by the radiation power projected onto the infrared sensor's sensitive element is exactly equal to the noise voltage of the sensor itself, then this radiation power is called "noise equivalent power". It is usually represented by the symbol "NEP".

[0113]

[0114] In the formula: Us is the output voltage of the infrared detector; P0 is the power projected onto the unit area of ​​the infrared sensitive element; A0 is the area of ​​the infrared sensitive element; UN is the comprehensive noise voltage of the infrared detector (generally the noise voltage); RV is the voltage response rate of the infrared detector.

[0115] (2) Calculate the detection rate D

[0116] The detection rate is the inverse of the noise equivalent power, that is,

[0117]

[0118] The higher the detection rate of an infrared sensor, the smaller the minimum radiation power that the sensor can detect, and the more sensitive the sensor is.

[0119] (3) Finally, calculate the detection ratio D*

[0120] Since the value of the detection rate D is related to the test conditions and the size and model of the detector, it is difficult to compare the detection rates D of different detectors. Therefore, the detection rate D is normalized to remove the differences caused by the test conditions and the size and model of the detector.

[0121] Extensive theoretical and experimental studies have shown that the signal-to-noise ratio of the detector output is proportional to the square root of the photosensitive surface area, that is, the detection rate is inversely proportional to the square root of the photosensitive element area. The total noise power spectrum of the detector is relatively flat in the mid-frequency band, and it can be considered that the noise voltage is only proportional to the square root of the measurement circuit bandwidth, that is, the detection rate is inversely proportional to the square root of the measurement circuit bandwidth. Therefore, a detection rate normalized to the photosensitive element area and the measurement circuit bandwidth can be defined, namely D*.

[0122]

[0123] D* is the normalized detection rate, called relative detection rate, and is read as D*. The physical meaning of D* is the signal-to-noise ratio obtained when 1W of radiation flux is incident on a detector with a photosensitive element area of ​​1cm2 and measured with a circuit with a bandwidth of 1Hz. The unit of D* is cm·Hz1 / 2·w-1. Using D* to compare the noise performance of two detectors can avoid the influence of different photosensitive element areas or measurement bandwidths on the detection rate measurement results.

[0124] (4) The detection ratio D* is the required value

[0125] 1.4 Response rate R

[0126] The output signal value generated by the system for each unit of incident radiation power is called the responsivity of the system.

[0127] In the actual system response rate test, it can be attributed to the response signal test of the imaging system under two black body irradiation conditions with different temperatures (T0, T). The response rate calculation formula of pixel i is as follows:

[0128]

[0129] Where DN(i,T0) and DN(i,T) are the pure response signal code values ​​of pixel i under the blackbody temperature T0 and T conditions, respectively, and are the average values ​​of the signal code values ​​obtained by using an infrared focal plane device to observe the blackbody with a temperature of T multiple times; and are the average spectral radiances received by pixel i under the blackbody temperatures T0 and T respectively, which are calculated by the spectral response function and the observed blackbody emissivity.

[0130] In the same detector, the response rates of different pixels should be as close as possible, so as to maximize the number of quantization bits of the A / D converter to achieve the requirements of high temperature resolution and high dynamic range. For multi-column long-line array focal plane detectors, the signals of each column of detector pixels are serially read out by a readout circuit. If the response rate of a pixel in the detector array is too high, it will appear as a saturated bright line in the image, and if the response rate is too low, it will appear as a dark line. All of these will affect the quantification quality of the data and the quality of the overall image.

[0131] 1.5 Dark background response DNbg

[0132] Dark background response DNbg refers to the response signal of the detector when there is no target incident energy or the incident energy is extremely low. In this case, the output signal of the instrument mainly comes from the background radiation of the instrument itself and the dark current signal of the detector. Similar to the response rate R, the dark background response DNbg of each pixel should be as close as possible, so as to maximize the quantization bit number of the A / D converter to achieve the requirements of high temperature resolution and high dynamic range.

[0133] The specific measurement method is to cover the infrared focal plane device or observe an extremely low temperature target to obtain the pure response signal code value DNbg in this case, which is what is required.

[0134] 1.6 Other indicators

[0135] As required, other indicators of the detector pixel measured, such as the response signal DN at a certain temperature, the response signal difference ΔDN at different temperatures, the detection rate D, the noise equivalent power NEP, etc., can also be uniformly compared using the method proposed in the present invention.

[0136] S2: preprocessing the data to be evaluated to filter bad pixels, and obtaining performance index data of each pixel after filtering;

[0137] Acquire the data to be evaluated and calculate the performance index of each pixel;

[0138] Identify the target pixels that meet any of the following index conditions as bad pixels and filter the bad pixels; it can be understood that after measuring and calculating the performance indicators of each pixel, the bad pixels in the infrared focal plane device should be removed. Bad pixels can be divided into blind pixels, overheated pixels, flickering pixels, etc. Pixels that meet any of the following conditions are bad pixels.

[0139] Wherein, the target pixels include:

[0140] When the performance indicator of the current pixel is the response rate R, the response rate R is less than 0.5 times the average response rate of the focal plane pixel;

[0141] When the performance indicator of the current pixel is the response rate R, the pixel whose response rate R is greater than 1.5 times the average response rate of the focal plane pixel;

[0142] When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is less than 0.5 times the average response value of the focal plane pixel;

[0143] When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is 1.5 times greater than the average response value of the focal plane pixel;

[0144] When the performance indicator of the current pixel is the noise voltage noise, the pixel whose noise voltage noise is greater than twice the average noise voltage of the focal plane pixel;

[0145] When the performance index of the current pixel is the noise equivalent temperature difference NEΔT, the pixel whose noise equivalent temperature difference NEΔT is greater than the preset threshold;

[0146] The bad pixels include blind pixels, overheated pixels, flickering pixels and related data that affect the performance and image quality of the image sensor.

[0147] When calculating the mean, standard deviation and z-score of each indicator later, bad pixels should be excluded.

[0148] S3: Calculate the mean and standard deviation of each performance index of all pixels after filtering. After removing bad pixels, calculate the mean E(index) and standard deviation σ(index) of each performance index of all pixels.

[0149] S4: Calculate the z score of the performance index corresponding to each pixel according to the average value and standard deviation of the performance index corresponding to each pixel, and summarize and add the z scores of each performance index corresponding to each pixel to obtain the final z score result of the comprehensive performance index of each pixel;

[0150] Calculate the z-score of each performance indicator for each pixel: The z-score, also known as the standard score or standard score, is a unitless symbolic number used in statistics to measure the standard deviation of the value of a certain observation point or data point above the mean of the observed or measured value. The z-score can unify indicators of different units and different scales for comparison. The z-score can evaluate the position or ranking of the value of a certain observation point or data point among all observation points or data points. The higher the z-score, the closer the observation point or data point is to the front among all observation points or data points. The closer the z-score is to 0, the closer the observation point or data point is to the middle among all observation points or data points. The smaller the z-score, the further back the observation point or data point is among all observation points or data points.

[0151] The calculation formula for the z-score zi of an observation point Xi or data point is as follows:

[0152]

[0153] Where Xi is the data of a certain observation point or data point, E(X) is the expected value of all observation points or data points, which is generally replaced by the arithmetic mean, and σ(X) is the standard deviation of all observation points or data points.

[0154] Calculating the z score of the performance indicator corresponding to each pixel according to the average value and standard deviation of the performance indicator corresponding to each pixel in step S4 includes:

[0155] The z-score calculation method for each performance index index of each pixel i is as follows:

[0156] For positive indicators, the z-score is:

[0157]

[0158] For negative indicators, the z-score is:

[0159]

[0160] For the uniformity index, the z-score is

[0161]

[0162] Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and 0 respectively. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, indexi The data of the pixel at the observation point or data point.

[0163] Calculate the z-score for each pixel: The final z-score of the comprehensive performance index for each pixel is calculated as follows:

[0164]

[0165] Among them, k(index) is the weight coefficient of each indicator, which is a non-negative number. The indicator weight coefficient k(index) can be set according to specific needs. For example, if you are more concerned about the detection ability of the pixel, you can set the weight coefficients of the noise voltage noise, the noise equivalent temperature difference NEΔT, and the detection rate D* higher; if you are more concerned about the uniformity of the pixel, you can set the weight coefficients of the detection rate R, the dark background response DNbg, and the spectral response function error higher. In the embodiment of the present invention, all weight coefficients are set to the same, all 100, for subsequent explanation.

[0166] S5: According to the final z-score result of each pixel, the pixel with the highest z-score in each row of the infrared focal plane array device is selected and registered as the optimal pixel. In this way, it is helpful to comprehensively consider the various indicators of each pixel, evaluate the quality of the pixel as a whole, and select the optimal pixel.

[0167] In another embodiment, a restriction range is set for each indicator, and pixels within the range are selected row by row. If two or more pixels in the current row meet the restriction range, the restriction range is narrowed until there is only one available pixel left in the current row. If there is no pixel meeting the restriction in the current row, the restriction is relaxed until only one pixel in all rows meets the requirement. This operation is repeated until the optimal pixels are selected for all rows.

[0168] S6: z-score normalization: After the z-score is calculated, in order to make the z-score more intuitive, the z-score can be normalized and linearly mapped to 0 to 1.

[0169]

[0170] In the formula represents the normalized z-score.

[0171] For example: Take the following data as an example to perform pixel optimization. The following data is for an infrared detector module, with a pixel size of 320 rows and 4 columns, numbered from pixel 1 to pixel 1280. The noise voltage noise, response rate R, detection rate D*, and noise correction temperature difference NEΔT have been measured for each pixel, and the indicators of each pixel are as follows:

[0172] Original pixel index data table 1:

[0173]

[0174]

[0175] After blind pixel elimination, the indicators are calculated as follows:

[0176] Average noise: 0.859

[0177] Response rate non-uniformity: 7.87%

[0178] Average detection rate D*: 1.11E+11

[0179] Average noise equivalent temperature difference: 60.8

[0180] The method of the embodiment of the present invention is used to select the best pixel, and the best pixel is selected in each row. The indicators of the selected pixels are as follows:

[0181] Pixel index data table 2 after selection in this embodiment:

[0182]

[0183]

[0184] The calculated indicators are:

[0185] Average noise: 0.661 (23% improvement)

[0186] Response rate non-uniformity: 5.2% (34% improvement)

[0187] Average detection rate D*: 1.385E+11 (improved by 25%)

[0188] Average Noise Equivalent Temperature Difference: 45.76 (25% improvement)

[0189] After using the pixel optimization method of the embodiment of the present invention, each index of the selected pixel is improved to a certain extent.

[0190] Embodiment 2

[0191] Based on the same inventive concept, the present invention provides an infrared detector focal plane pixel evaluation and selection device based on z-score, comprising:

[0192] A data acquisition module, used for acquiring the data to be evaluated of each pixel in the infrared focal plane array device;

[0193] A preprocessing module, used for preprocessing the data to be evaluated to filter out bad pixels, and obtaining performance index data of each pixel after filtering;

[0194] The first calculation module is used to calculate the average value and standard deviation of each performance index of all pixels after filtering;

[0195] The second calculation module is used to calculate the z score of the performance index corresponding to each pixel according to the average value and standard deviation of the performance index corresponding to each pixel, and summarize and add the z scores of each performance index corresponding to each pixel to obtain the final z score result of the comprehensive performance index of each pixel;

[0196] A registration module is selected to select the pixel with the highest z score in each row of the infrared focal plane array device according to the final z score result of each pixel, and register it as the optimal pixel.

[0197] The second calculation module includes:

[0198] The z-score calculation method for each performance index index of each pixel i is as follows:

[0199] For positive indicators, the z-score is:

[0200]

[0201] For negative indicators, the z-score is:

[0202]

[0203] For the uniformity index, the z-score is

[0204]

[0205] Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and 0 respectively. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, index i The data of the pixel at the observation point or data point.

[0206] It should be noted that the division of the modules in the embodiment of the device / system is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or physically separated. Moreover, these modules can be implemented in the form of software calling through processing elements; or in the form of hardware; or some units can be implemented in the form of software calling through processing elements, and some units can be implemented in the form of hardware.

[0207] The implementation principles of the above modules have been described in the above embodiments, so they will not be repeated here.

[0208] Embodiment 3

[0209] Based on the same inventive concept, in some embodiments of the present application, an electronic device is also provided. The electronic device includes a memory and a processor, wherein the memory is used to store a processing program, and the processor executes the processing program according to instructions. When the processor executes the processing program, the infrared detector focal plane pixel evaluation and selection method based on the z score in the above-mentioned embodiment is implemented.

[0210] In some embodiments of the present application, a readable storage medium is further provided, which may be a non-volatile readable storage medium or a volatile readable storage medium. The readable storage medium stores instructions, and when the instructions are executed on a computer, an electronic device including the readable storage medium executes the aforementioned infrared detector focal plane pixel evaluation and selection method based on z-score.

[0211] It can be understood that, for the aforementioned method for evaluating and selecting infrared detector focal plane pixels based on z-scores, if they are all implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0212] Computer readable storage media may include data signals propagated in baseband or as part of a carrier wave, wherein readable program codes are carried. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, device, or device. The program codes contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0213] The program code involved in executing the technical solution disclosed in this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Python, C++, etc., and also conventional procedural programming languages ​​such as C language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0214] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating and selecting infrared detector focal plane pixels based on z-score, characterized in that: include: Acquire the data to be evaluated of each pixel in the infrared focal plane array device; Preprocessing the data to be evaluated to filter bad pixels, and obtaining performance index data of each pixel after filtering; Calculate the mean and standard deviation of each performance indicator for all pixels after filtering; The z score of the performance index corresponding to each pixel is calculated according to the mean value and standard deviation of the performance index corresponding to each pixel, and the z scores of each performance index corresponding to each pixel are summarized and added to obtain the final z score result of the comprehensive performance index of each pixel; According to the final z-score result of each pixel, the pixel with the highest z-score in each row of the infrared focal plane array device is selected and aligned as the optimal pixel.

2. The method for evaluating and selecting infrared detector focal plane pixels based on z-score according to claim 1, characterized in that: The data to be evaluated include the noise voltage noise, noise equivalent temperature difference NEΔT, specific detection rate D*, response rate R, signal voltage change ΔV, dark background response DNbg of each pixel and performance indicators related to different test conditions.

3. The method for evaluating and selecting infrared detector focal plane pixels based on z-score according to claim 2, characterized in that: The step of obtaining the data to be evaluated for each pixel in the infrared focal plane array device comprises: Based on the focal plane array device, observe the black body and record the current temperature T1, obtain the effective response value of each pixel, and calculate the noise root mean square noise1 at the black body temperature; Based on the focal plane array device, observe the black body and record the current temperature T2, obtain the effective response value of each pixel, and calculate the noise root mean square noise2 at the black body temperature; Calculate the noise equivalent temperature difference NEΔT: Among them, DN1 and DN2 are both valid response values.

4. The method for evaluating and selecting infrared detector focal plane pixels based on z-score according to claim 1, characterized in that: The preprocessing of the data to be evaluated to filter bad pixels, and obtaining the performance index data of each pixel after filtering includes: Acquire the data to be evaluated and calculate the performance index of each pixel; Identify target pixels that meet any of the following index conditions as bad pixels and filter out the bad pixels; Wherein, the target pixel includes: When the performance indicator of the current pixel is the response rate R, the response rate R is less than 0.5 times the average response rate of the focal plane pixel; When the performance indicator of the current pixel is the response rate R, the pixel whose response rate R is greater than 1.5 times the average response rate of the focal plane pixel; When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is less than 0.5 times the average response value of the focal plane pixel; When the performance index of the current pixel is the response value DN, at the same measurement temperature, the pixel whose response value DN is 1.5 times greater than the average response value of the focal plane pixel; When the performance indicator of the current pixel is the noise voltage noise, the pixel whose noise voltage noise is greater than twice the average noise voltage of the focal plane pixel; When the performance index of the current pixel is the noise equivalent temperature difference NEΔT, the pixel whose noise equivalent temperature difference NEΔT is greater than the preset threshold; The bad pixels include blind pixels, overheated pixels, flickering pixels and related data that affect the performance and image quality of the image sensor.

5. The method for evaluating and selecting infrared detector focal plane pixels based on z-score according to claim 1, characterized in that: Calculating the z score of the performance indicator corresponding to each pixel according to the average value and standard deviation of the performance indicator corresponding to each pixel includes: The z-score calculation method for each performance index index of each pixel i is as follows: For positive indicators, the z-score is: For negative indicators, the z-score is: For the uniformity index, the z-score is Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and 0 respectively. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, index i The data of the pixel at the observation point or data point.

6. The method for evaluating and selecting infrared detector focal plane pixels based on z-score according to claim 5, characterized in that: The z-score of each performance indicator corresponding to each pixel is summarized and added to obtain the final z-score result of the comprehensive performance indicator of each pixel, including: The final z-score for the overall performance metric for each pixel is calculated as follows: Among them, k(index) is the weight coefficient of each indicator.

7. An infrared detector focal plane pixel evaluation and selection device based on z-score, characterized in that: include: A data acquisition module, used for acquiring the data to be evaluated of each pixel in the infrared focal plane array device; A preprocessing module, used for preprocessing the data to be evaluated to filter out bad pixels, and obtaining performance index data of each pixel after filtering; The first calculation module is used to calculate the average value and standard deviation of each performance index of all pixels after filtering; The second calculation module is used to calculate the z score of the performance index corresponding to each pixel according to the average value and standard deviation of the performance index corresponding to each pixel, and summarize and add the z scores of each performance index corresponding to each pixel to obtain the final z score result of the comprehensive performance index of each pixel; A registration module is selected to select the pixel with the highest z score in each row of the infrared focal plane array device according to the final z score result of each pixel, and register it as the optimal pixel.

8. The infrared detector focal plane pixel evaluation and selection device based on z-score according to claim 7, characterized in that: The second calculation module includes: The z-score calculation method for each performance index index of each pixel i is as follows: For positive indicators, the z-score is: For negative indicators, the z-score is: For the uniformity index, the z-score is Among them, the z score of the i-th pixel under the index index is recorded as s represents different indicator types, which can be +, -, and 0 respectively. When S is +, Applicable to positive indicators; when s is -, Applicable to negative indicators, when s is 0, Applicable to uniformity index; E(index) is the average value of the pixel under the index index, σ(index) is the standard deviation of the pixel under the index index, index i The data of the pixel at the observation point or data point.

9. A computer device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the infrared detector focal plane pixel evaluation and selection method based on z-score as described in any one of claims 1 to 6.

10. An electronic device, characterized in that: include: A memory, the memory being used to store a processing program; A processor, wherein when executing the processing program, the processor implements the infrared detector focal plane pixel evaluation and selection method based on z-score as described in any one of claims 1 to 6.

Images