A non-cooled long-wave infrared camera multivariate fitting temperature calibration method and computer program product
Patent Information
- Application Number
- CN202510467374.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-14
- Publication Date
- 2026-10-20
AI Technical Summary
[0006]本发明的目的是解决现有技术中非制冷长波红外相机的温度标定方法受探测器温度漂移、光学镜头温度变化引起杂散辐射的影响,测温精度难以保证的技术问题,而提供一种非制冷长波红外相机多变量拟合温度标定方法、计算机程序产品,用相机输出灰度、镜头温度、探测器焦平面温度这三个变量进行建模拟合,大幅提升了基于氧化钒红外探测器的非制冷长波红外相机测温精度
[0047]本发明的一种非制冷长波红外相机多变量拟合温度标定方法、计算机程序产品,通过采用相机输出灰度、镜头温度、探测器焦平面温度这三个变量进行建模拟合的方式,大幅提升了基于氧化钒红外探测器的非制冷长波红外相机测温精度;拟合的因变量选择辐亮度而非温度,实现了更好的线性度,有利于降低过拟合风险和误差。
Smart Images

Figure BDA0005358900040000041 
Figure BDA0005358900040000081 
Figure BDA0005358900040000082
Abstract
Description
Technical Field
[0001] This invention relates to a camera temperature calibration method, specifically to a multivariable fitting temperature calibration method and computer program product for an uncooled long-wave infrared camera. Background Technology
[0002] Non-contact temperature measurement plays an irreplaceable role in national defense and epidemic prevention. Among them, uncooled long-wave infrared cameras, which can combine imaging and temperature measurement capabilities, are widely used in national defense, power industry, civilian security, and epidemic prevention. High-precision temperature radiation calibration and inversion are among the key technologies that determine the temperature measurement performance of infrared cameras, and are highly correlated with the accuracy of temperature measurement.
[0003] Infrared detectors are the core components of infrared cameras. According to research, long-wavelength uncooled infrared detectors belong to the category of microthermometers, with vanadium oxide (VOx) or amorphous silicon (α-Si) as their core thermistor materials. Currently, the mainstream domestically produced long-wavelength uncooled infrared detectors on the Chinese market are all vanadium oxide detectors. In recent years, to achieve miniaturization, low cost, and low power consumption applications, most vanadium oxide long-wavelength infrared detectors have adopted ceramic packaging and do not integrate a thermoelectric cooler (TEC) module. Therefore, they cannot maintain the focal plane of the infrared detector at a stable temperature point. The thermoelectric properties of vanadium oxide material change with temperature, affecting its infrared spectral transmittance. For vanadium oxide long-wavelength infrared detectors, when operating at room temperature, the temperature of the detector's focal plane gradually rises until thermal equilibrium is reached as the operating time increases. During this process, if a constant-temperature blackbody is placed directly in front of the detector, the grayscale of the original image output by the detector will gradually decrease with increasing operating temperature. This characteristic results in a large calibration error for infrared cameras using vanadium oxide detectors for temperature measurement. Furthermore, according to Planck's law and Wien's displacement law, the radiation peak of the temperature range in which the camera operates is in the long-wavelength band (e.g., the radiation peak of a blackbody at room temperature of 25°C is 9.7 μm), which falls precisely within the response band (8 μm to 14 μm) of the vanadium oxide long-wavelength infrared detector. Therefore, the detector is inevitably affected by stray radiation caused by the temperature rise of the lens. In applications such as optoelectronic pods, long-wavelength infrared cores are often placed in enclosed spaces, resulting in significant temperature rises and difficulty in achieving thermal equilibrium, making it difficult to guarantee the accuracy of radiation calibration.
[0004] Most existing methods typically involve fixing a blackbody in front of an infrared camera, acquiring output images at different blackbody temperatures, calculating the average gray level G of the area covered by the blackbody, and then fitting the image using gray level as the independent variable and temperature T as the dependent variable to obtain a fitting formula. In temperature measurement applications, measurements are taken on a blackbody or target object (with known or defined emissivity) to obtain the grayscale value of a local area of the target. This grayscale value is then substituted into a fitting formula to calculate the target temperature. However, this method is significantly affected by factors such as temperature drift of the vanadium oxide detector and stray radiation changes caused by variations in the temperature of the optical lens, making it difficult to guarantee measurement accuracy. Currently, mainstream infrared temperature measurement products in China typically have a temperature measurement accuracy of ±1℃ for the 0℃~150℃ range, which cannot meet the requirements of some high-precision applications.
[0005] In conclusion, researching high-precision temperature calibration methods for uncooled long-wave infrared cameras is crucial for improving the temperature measurement accuracy of uncooled long-wave infrared cameras in closed systems such as optoelectronic pods, and has significant research significance and practical value. Summary of the Invention
[0006] The purpose of this invention is to solve the technical problem that the temperature calibration method for uncooled long-wave infrared cameras in the prior art is affected by stray radiation caused by detector temperature drift and optical lens temperature changes, making it difficult to guarantee temperature measurement accuracy. The invention provides a multivariate fitting temperature calibration method and computer program product for uncooled long-wave infrared cameras, which uses three variables—camera output grayscale, lens temperature, and detector focal plane temperature—for modeling and fitting, significantly improving the temperature measurement accuracy of uncooled long-wave infrared cameras based on vanadium oxide infrared detectors.
[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:
[0008] A method for multivariate fitting temperature calibration of an uncooled long-wave infrared camera, characterized by the following steps:
[0009] Step 1: Place the blackbody within the field of view of the infrared camera and test it to ensure that the blackbody completely covers the field of view of the uncooled long-wave infrared camera.
[0010] Step 2: Set the temperature measurement range of the uncooled long-wave infrared camera, and set m temperature calibration points according to the temperature measurement range. Then, use the control computer to collect the blackbody temperature T at each temperature calibration point. b Image data;
[0011] Step 3: Perform calibration data preprocessing on the acquired image data, and obtain the number of calibration data points n1, n2, n3, ..., n from the m temperature calibration points respectively. m We obtain equal-length vectors: detector temperature Td, lens temperature T l Mean G of blackbody image b and blackbody radiance R b ;
[0012] Step 4: Using MATLAB software, construct the polynomial matrix Array_fitting, and obtain the regression coefficient b_regression based on the equal-length vector from Step 3.
[0013] Step 5: Invert the blackbody temperature T′ using the regression coefficient b_regression. b If the blackbody temperature T′ is inverted b With blackbody temperature T b The difference between them is within ±1 degree, and the inverted blackbody temperature T′ is obtained. b This is the measured temperature value of the uncooled long-wave infrared camera, completing the calibration; if the difference is greater than ±1 degree, check and adjust the temperature accuracy of the blackbody and the stability of the camera's response, and return to step 3 until the difference is within ±1 degree, completing the calibration.
[0014] Furthermore, the specific steps of step 2 are as follows:
[0015] Step 2.1: Set the temperature measurement range of the uncooled long-wave infrared camera to 0℃~500℃;
[0016] Step 2.2, set the temperature calibration point T c =[T c1 T c2 T c3 …, T cm ];
[0017] Step 2.3, set the blackbody temperature T b Set as the first temperature calibration point T c1 Keeping the uncooled long-wave infrared camera and the blackbody in a fixed position, one frame of image data is acquired every minute; when the uncooled long-wave infrared camera reaches thermal equilibrium, the first temperature calibration point T is completed. c1 Image data acquisition;
[0018] Step 2.4, sequentially change the blackbody temperature T b Set to T respectively c2 T c3 ...... T cm Following the method in step 2.3, complete the image data acquisition for all temperature calibration points.
[0019] Furthermore, step 3 specifically includes:
[0020] For temperature calibration point T c1 T c2 ... T cm The corresponding radiance R is calculated using the following formula. b1 R b2 ..., R bm ;
[0021]
[0022] Where ε is the emissivity of the blackbody or the target under test; λ1 is the lower limit of the response band of the uncooled long-wave infrared camera; λ2 is the upper limit of the response band of the uncooled long-wave infrared camera; λ is the wavelength of the uncooled long-wave infrared camera; C1 is the first thermodynamic radiation constant; C2 is the second thermodynamic radiation constant; and T is the thermodynamic temperature corresponding to each temperature calibration point.
[0023] Furthermore, the detector temperature Td and the lens temperature T l Mean G of blackbody image b and blackbody radiance R b The set is:
[0024] T d =[T d_1_1 T d_1_2 ,…,T d_1_n1 T d_2_1 T d_2_2 ,…,T d_2_n2 ,…,T d_m_1 ,…,T d_m_nm ]
[0025] T l =[T l_1_1 T l_1_2 ,…,T l_1_n1 T l_2_1 T l_2_2 ,…,T l_2_n2 ,…,T l_m_1 ,…,T l_m_nm ]
[0026] G b =[G b_1_1 G b_1_2 , ..., G b_1_n1 G b_2_1 G b_2_2 , ..., G b_2_n2 , ..., G b_m_1 , ..., G b_m_nm ]
[0027] R b =[R b_1_1 R b_1_2 , ..., R b_1_n1 R b_2_1 R b_2_2 , ..., R b_2_n2 , ..., R b_m_1 , ..., R b_m_nm ]
[0028] Among them, R b_1_1 Rb_1_2 , ..., R b_1_n1 This represents the blackbody radiance corresponding to the n1th temperature calibration point;
[0029] R b_m_1 , ..., R b_m_nm For the nth m Blackbody radiance corresponding to each temperature calibration point;
[0030] G b_1_1 G b_1_2 , ..., G b_1_n1 This is the mean value of the blackbody image corresponding to the n1th temperature calibration point;
[0031] G b_m_1 , ..., G b_m_nm For the nth m Mean value of the blackbody image corresponding to each temperature calibration point;
[0032] T l_1_1 T l_1_2 ,…,T l_1_n1 This is the lens temperature corresponding to the n1th temperature calibration point;
[0033] T l_m_1 ,…,T l_m_nm For the nth m The lens temperature corresponding to each temperature calibration point;
[0034] T d_1_1 T d_1_2 ,…,T d_1_n1 The detector temperature corresponding to the n1th temperature calibration point;
[0035] T d_m_1 ,…,T d_m_nm Represents the nth m The detector temperature corresponding to each temperature calibration point.
[0036] Furthermore, in step 5, the blackbody temperature T′ b Through blackbody radiance R′ b The calculated blackbody radiance R′ b The steps to obtain it are as follows:
[0037] Step a.1, set the blackbody temperature T b The temperature set as the non-calibration point, combined with the regression coefficient b_regression, is used to obtain the detector temperature T′ at the non-calibration temperature according to steps 2 and 3. d Lens temperature T′ l Mean G′ of blackbody image b ;
[0038] Step a.2, set the detector temperature T′ dLens temperature T′ l Mean G′ of blackbody image b The vector is converted and substituted into the polynomial matrix Array_fitting constructed in step 4 for calculation, resulting in the matrix Array_test;
[0039] Step a.3, calculate the inverted blackbody radiance R′ using the following formula. b :
[0040] R′ b =Array test ×b_regression.
[0041] Furthermore, in step 4, the regression coefficients are specifically:
[0042] b_regression=regress(R b (Array_fitting)
[0043] Here, `regress` is the function in MATLAB software for calculating regression coefficients.
[0044] Furthermore, in step 3, ε takes a value of 0.95 to 0.99; λ1 is 8 μm; λ2 is 14 μm; and C1 is 3.741772e -16 C2 is 0.014388.
[0045] A computer program product comprising a computer program, characterized in that: when executed by a processor, the program implements the steps of the multivariable fitting temperature calibration method for an uncooled long-wave infrared camera as described in any one of claims 1 to 7.
[0046] Compared with the prior art, the present invention has the following beneficial technical effects:
[0047] The present invention discloses a multivariate fitting temperature calibration method and computer program product for an uncooled long-wave infrared camera. By using three variables—camera output grayscale, lens temperature, and detector focal plane temperature—for modeling and fitting, the temperature measurement accuracy of an uncooled long-wave infrared camera based on a vanadium oxide infrared detector is significantly improved. The dependent variable for fitting is radiance rather than temperature, which achieves better linearity and helps reduce the risk of overfitting and errors. Attached Figure Description
[0048] Figure 1 This is a schematic diagram of the test system connection in an embodiment of the multivariable fitting temperature calibration method for an uncooled long-wave infrared camera of the present invention;
[0049] Figure 2 This is a flowchart illustrating an embodiment of the multivariable fitting temperature calibration method for an uncooled long-wave infrared camera according to the present invention; Detailed Implementation
[0050] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.
[0051] The readout circuit of the vanadium oxide infrared detector can obtain the real-time operating temperature of the focal plane of the vanadium oxide detector (hereinafter referred to as "detector temperature"), which is defined as Td; by attaching a surface-mount temperature sensor to the optical lens, the real-time temperature of the lens (hereinafter referred to as "lens temperature") is obtained, which is defined as T. l The true value of the blackbody temperature (hereinafter referred to as "blackbody temperature") during the acquisition and calibration data is defined as T. b The image acquired by an uncooled long-wave infrared camera when calibrated against a blackbody is defined as I. b The gray-scale mean of the effective region of a blackbody target in an image (hereinafter referred to as "blackbody image mean") is defined as G. b The interconnection between the blackbody, the uncooled long-wave infrared camera, and the computer acquiring the data is as follows: Figure 1 As shown.
[0052] Multiple sets of data were collected, corresponding to different blackbody temperatures (data could only be collected after the blackbody temperature had stabilized each time), as well as different detector and lens temperatures, forming a data vector. The detector temperature Td and lens temperature T were defined. l Mean G of blackbody image b As the independent variable, the radiance R corresponding to the blackbody temperature b Let be the dependent variable, and the correspondence between the independent variable and the dependent variable is shown in formula (1).
[0053] R b =f(T) d ,T l G b (1)
[0054] Construct a suitable polynomial model, then input calibration data for fitting, and obtain the coefficients of each polynomial of the function f in formula (1). At this point, calibration and fitting are complete, and it can be used for temperature measurement applications. For a blackbody or target object with known emissivity, use a similar... Figure 1 Image data is acquired using a specific method, the average grayscale value of the target area is calculated, and the detector temperature and lens temperature recorded during data acquisition are input to retrieve the surface temperature of the target. This invention provides temperature measurement accuracy better than ±1℃ for blackbodies or targets with known emissivity ranging from 0℃ to 150℃, and better than ±2.5℃ for blackbodies or targets with known emissivity ranging from 0℃ to 500℃.
[0055] like Figure 2 As shown, the multivariable fitting temperature calibration method for an uncooled long-wave infrared camera of the present invention specifically includes the following steps:
[0056] The first step is equipment setup and connection. Place the blackbody, uncooled long-wave infrared camera, computer, and other equipment as needed. Figure 1 As shown; the detector temperature is defined as T. d The lens temperature is T. l The blackbody temperature is T b (The units for all three temperatures are °C), and the mean value of the blackbody image is G. b After setting up the equipment, testing is required to ensure that the blackbody covers the infrared camera's field of view. During calibration, the indoor ambient temperature needs to be kept relatively stable (a stable ambient temperature is not required for verification and temperature measurement applications).
[0057] The second step is to set appropriate temperature calibration points based on the camera's temperature measurement range requirements. Define the lowest temperature for camera temperature measurement as T. min (Generally 0℃), the highest temperature is T max (Generally greater than 100℃), the required calibrated temperature point T c =[T c1 T c2 T c3 …, T cm The value can be calculated according to formula (2). (Since calibration below 0℃ requires a low-temperature cavity blackbody, its application is relatively rare. This invention mainly targets high-precision calibration and inversion within the temperature range of 0℃ to 500℃. Furthermore, the uncooled long-wave infrared thermometer is in roller integration mode with only one integration time, and the integration time will not be switched during operation. The setting of this integration time can also cover the 0℃ to 500℃ range.) The method of this invention is only applicable to the following two temperature ranges: up to 5 calibration points within 150 degrees; and 8 calibration points between 150 and 500 degrees.
[0058]
[0059] The third step is to complete the image data acquisition for the first temperature calibration point. Turn on the uncooled long-wave infrared camera, computer, blackbody, and other equipment, and set the blackbody temperature to the first calibration point T. c1 Keeping all equipment positions fixed, once the blackbody temperature stabilizes, acquire one frame every minute. During this period, the detector temperature T... d and lens temperature T l Gradually increase the temperature from room temperature to around 50°C, maintaining a data acquisition rate of 1 frame per minute, until the camera reaches thermal equilibrium. At this point, T... d and T lThe temperature remains relatively stable. After data acquisition is complete, all devices are turned off and allowed to cool to room temperature. The principle assumes a thermal equilibrium (Tc) of 30°C and an ambient temperature of 15°C. Initially, the lens temperature is around 15°C, and the detector temperature will be slightly higher (due to heat generation after startup). As the camera operates for longer, the heat generated causes the detector and lens temperatures to gradually increase. Generally, they reach equilibrium when they are more than 30°C higher than the ambient temperature. For example, if the ambient temperature remains stable at 15°C, the camera temperature will rise to approximately 45°C and then remain relatively stable, indicating thermal equilibrium. No further data acquisition is necessary.
[0060] The fourth step is to complete the image data acquisition for the other calibration temperature points. The blackbody temperature is set sequentially to T. c2 T c3 Once the calibration temperature points are reached, repeat step three until all temperature calibration point data collection is completed.
[0061] The fifth step is data preprocessing. Using the `mean` function in MATLAB software, the mean value G of the blackbody image is calculated for all acquired image data. b The calculation method involves taking the global mean of the image (using the mean function). Assuming there are m temperature calibration points, for temperature calibration point T... c1 T c2 ... T cm The corresponding radiance R is calculated using formula (3). b1 R b2 ..., R bm .
[0062]
[0063] The variables in formula (3) are as follows:
[0064] ε represents the emissivity of the blackbody or the target being measured. The emissivity of the blackbody needs to be found in the equipment manual, while the emissivity of the target being measured is generally provided by the user or obtained from a table of emissivity values for common objects. For example, the emissivity of water is generally 0.95, and the emissivity of a blackbody is usually between 0.95 and 0.99. λ is the wavelength of the uncooled long-wave infrared camera.
[0065] λ1 represents the lower limit of the response band of an infrared camera. The lower limit of the response band of a long-wave uncooled infrared camera is usually 8μm.
[0066] λ2 represents the upper limit of the response band of an infrared camera. The upper limit of the response band of a long-wave uncooled infrared camera is usually 14μm.
[0067] C1 represents the first thermodynamic radiation constant, with a value of C1 = 3.741772e -16 ;
[0068] C2 represents the second thermodynamic radiation constant, with a value of C2 = 0.014388;
[0069] T represents thermodynamic temperature (unit: Kelvin / K), which is the actual temperature value at a given point (unit: °C). c +273.15; For example, if the blackbody temperature is set to 20℃, then the thermodynamic temperature is 293.15K, and then 293.15K is substituted into the formula to calculate the radiance.
[0070] Step 6: Preparation of fitting data. The number of calibration data points for each temperature calibration point are n1, n2, n3, ..., n m This can form several equal-length vectors as shown in formulas (4) to (7), representing the detector temperature T respectively. d Lens temperature T l Mean G of blackbody image b Blackbody radiance R b , where R b_1_1 R b_1_2 , ..., R b_1_n1 Representing the blackbody radiance corresponding to the same temperature calibration point, all values are R. c1 And so on for the others.
[0071] T d =[T d_1_1 T d_1_2 ,…,T d_1_n1 T d_2_1 T d_2_2 ,…,T d_2_n2 ,…,T d_m_1 ,…,T d_m_nm (4)
[0072] T l =[T l_1_1 T l_1_2 ,…,T l_1_n1 T l_2_1 T l_2_2 ,…,T l_2_n2 ,…,T l_m_1 ,…,T l_m_nm (5)
[0073] G b =[G b_1_1 G b_1_2 G b_1_n1 G b_2_1 G b_2_2 G b_2_n2 G b_m_1 , ···, T b_m_nm (6)
[0074] Rb =[R b_1_1 R b_1_2 , ..., R b_1_n1 R b_2_1 R b_2_2 , ..., R b_2_n2 , ..., R b_m_1 , ..., R b_m_nm (7)
[0075] Assuming the blackbody temperature calibration points are set to 30℃, 60℃, 90℃, 120℃, and 150℃, corresponding to 1 to m calibration points (m=5), where the first calibration point collected 15 sets of data, the second collected 17 sets, the third collected 20 sets, the fourth collected 18 sets, and the fifth collected 15 sets, then n1=15, n2=17, n3=20, n4=18, and n5=15. This n5 is also known as nm.
[0076] R b To determine the blackbody radiance, m calibration points (defined as 1 to m) are set at the blackbody temperature. The calibration data for each point are n1, n2, n3 to nm (the data volume for each calibration point is not necessarily equal because the time from calibration to thermal equilibrium is not exactly the same). The annotations following the calibration settings are: b_l_1 represents the first set of calibration data collected at the first set calibration temperature point of the blackbody; b_1_2 represents the second set of calibration data collected at the first set temperature calibration point of the blackbody; b_m_1 represents the first set of calibration data collected at the m-th set temperature calibration point of the blackbody; and b_m_nm represents the nm-th set of calibration data collected at the m-th set temperature calibration point of the blackbody.
[0077] Step 7: Data fitting calibration. Using MATLAB software, construct the polynomial matrix Array_fitting, as shown in Equation (8). Then calculate the regression coefficient b_regression, as shown in Equation (9). In Equation (8), ones and length are the matrix and measurement vector length functions of MATLAB software, and in Equation (9), regression is the regression coefficient calculation function of MATLAB software.
[0078]
[0079] b_regression=regress(R b ,Array_fitting) (9)
[0080] Step 8: Calibration-Inversion Accuracy Verification. The blackbody temperature is set to a temperature outside the calibration point, such as T. test1 =(Tc1 +T c2 ) / 2、T test2 =(T c2 +T c3 Repeat step 3 (e.g., ) / 2) to complete data acquisition. Then repeat step 5 to complete data preprocessing. Finally, calculate the detector temperature T′. d Lens temperature T′ l Mean G′ of blackbody image b Substituting into formula (8), we obtain the matrix Array_test. Then, according to formula (10), we calculate the inverted blackbody radiance R′. b .
[0081] R′ b =Array test ×b_regression (10)
[0082] Next, the blackbody radiance R′ b Substituting into the inverse function of formula (3) to calculate the blackbody temperature T′ b Finally, based on the inverted blackbody temperature T′ b Compared with the actual blackbody temperature T b The difference between the values is used to evaluate the calibration-inversion accuracy. In practical temperature measurement applications, the inversion of the blackbody temperature T′ is used. b This refers to the measured temperature value of the target object.
[0083] The uncooled long-wave infrared camera multivariable fitting temperature calibration method of the present invention can also be formed into a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the uncooled long-wave infrared camera multivariable fitting temperature calibration method.
[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.
Claims
1. A method for multivariable fitting temperature calibration of an uncooled long-wave infrared camera, characterized in that, Includes the following steps: Step 1: Place the blackbody within the field of view of the infrared camera and test it to ensure that the blackbody completely covers the field of view of the uncooled long-wave infrared camera. Step 2: Set the temperature measurement range of the uncooled long-wave infrared camera, set m temperature calibration points according to the temperature measurement range, and collect the image data of the blackbody temperature Tb of each temperature calibration point by controlling the computer. Step 3: Perform calibration data preprocessing on the acquired image data, and obtain the number of calibration data points n1, n2, n3, ..., n from the m temperature calibration points respectively. m We obtain equal-length vectors: detector temperature Td, lens temperature T l Mean G of blackbody image b And blackbody radiance Rb; Step 4: Using MATLAB software, construct the polynomial matrix Array_fitting, and obtain the regression coefficient b_regression based on the equal-length vector from Step 3. Step 5: Invert the blackbody temperature T′ using the regression coefficient b_regression. b If the blackbody temperature T′ is inverted b The difference between the blackbody temperature Tb and the blackbody temperature T′ is within ±1 degree. b This is the measured temperature value of the uncooled long-wave infrared camera, completing the calibration; if the difference is greater than ±1 degree, check and adjust the temperature accuracy of the blackbody and the stability of the camera's response, and return to step 3 until the difference is within ±1 degree, completing the calibration.
2. The method for multivariable fitting temperature calibration of an uncooled long-wave infrared camera according to claim 1, characterized in that, The specific steps of step 2 are as follows: Step 2.1: Set the temperature measurement range of the uncooled long-wave infrared camera to 0℃~500℃; Step 2.2, set the temperature calibration point T c =[T c1 T c2 T c3 …, T cm ]; Step 2.3, set the blackbody temperature T b Set as the first temperature calibration point T c1 Keeping the uncooled long-wave infrared camera and the blackbody in a fixed position, one frame of image data is acquired every minute; when the uncooled long-wave infrared camera reaches thermal equilibrium, the first temperature calibration point T is completed. c1 Image data acquisition; Step 2.4, sequentially change the blackbody temperature T b Set to T respectively c2 T c3 ...... T cm Following the method in step 2.3, complete the image data acquisition for all temperature calibration points.
3. The method for multivariable fitting temperature calibration of an uncooled long-wave infrared camera according to claim 2, characterized in that, Step 3 specifically involves: For temperature calibration point T c1 T c2 ... T cm The corresponding radiance R is calculated using the following formula. b1 R b2 ..., R bm ; Where ε is the emissivity of the blackbody or the target under test; λ1 is the lower limit of the response band of the uncooled long-wave infrared camera; λ2 is the upper limit of the response band of the uncooled long-wave infrared camera; λ is the wavelength of the uncooled long-wave infrared camera; C1 is the first thermodynamic radiation constant; C2 is the second thermodynamic radiation constant; and T is the thermodynamic temperature corresponding to each temperature calibration point.
4. The multivariable fitting temperature calibration method for an uncooled long-wave infrared camera according to claim 3, characterized in that: In step 3, the detector temperature Td and the lens temperature T l Mean G of blackbody image b And the set of blackbody radiance Rb is: T d =[T d_1_1 ,T d_1_2 ,…,T d_1_n1 ,T d_2_1 ,T d_2_2 ,…,T d_2_n2 ,…,T d_m_1 ,…,T d_m_nm ] T l =[T l_1_1 ,T l_1_2 ,…,T l_1_n1 ,T l_2_1 ,T l_2_2 ,…,T l_2_n2 ,…,T l_m_1 ,…,T l_m_nm ] G b =[G b_1_1 ,G b_1_2 ,…,G b_1_n1 ,G b_2_1 ,G b_2_2 ,…,G b_2_n2 ,…,G b_m_1 ,…,G b_m_nm ] R b =[R b_1_1 ,R b_1_2 ,…,R b_1_n1 ,R b_2_1 ,R b_2_2 ,…,R b_2_n2 ,…,R b_m_1 ,…,R b_m_nm ] Among them, R b_1_1 R b_1_2 , ..., R b_1_n1 The blackbody radiance corresponding to the n1th temperature calibration point; R b_m_1 , ..., R b_m_nm For the nth m Blackbody radiance corresponding to each temperature calibration point; G b_1_1 G b_1_2 , ..., G b_1_n1 This is the mean value of the blackbody image corresponding to the n1th temperature calibration point; G b_m_1 , ..., G b_m_nm For the nth m Mean value of the blackbody image corresponding to each temperature calibration point; T l_1_ 1, T l_1_2 ,…,T l_1_n1 This is the lens temperature corresponding to the n1th temperature calibration point; T l_m_1 ,…,T l_m_nm For the nth m The lens temperature corresponding to each temperature calibration point; T d_1_1 T d_1_2 ,…,T d_1_n1 The detector temperature corresponding to the n1th temperature calibration point; T d_m_1 ,…,T d_m_nm Represents the nth m The detector temperature corresponding to each temperature calibration point.
5. The multivariable fitting temperature calibration method for an uncooled long-wave infrared camera according to claim 4, characterized in that, In step 5, the blackbody temperature T′ b Through blackbody radiance R′ b The calculated blackbody radiance R′ b The steps to obtain it are as follows: Step a.1, set the blackbody temperature T b The temperature set as the non-calibration point, combined with the regression coefficient b_regression, is used to obtain the detector temperature T′ at the non-calibration temperature according to steps 2 and 3. d Lens temperature T′ l Mean G′ of blackbody image b ; Step a.2, set the detector temperature T′ d Lens temperature T′ l Mean G′ of blackbody image b The vector is converted and substituted into the polynomial matrix Array_fitting constructed in step 4 for calculation, resulting in the matrix Array_test; Step a.3, calculate the inverted blackbody radiance R′ using the following formula. b : R′ b =Array test ×b_regression。 6. The multivariable fitting temperature calibration method for an uncooled long-wave infrared camera according to claim 5, characterized in that: In step 4, the regression coefficients are specifically: b_regression=regress(R b ,Array_fitting) Here, `regress` is the function in MATLAB software for calculating regression coefficients.
7. The multivariable fitting temperature calibration method for an uncooled long-wave infrared camera according to claim 6, characterized in that: In step 3, ε is set to 0.95–0.99; λ1 is 8 μm; λ2 is 14 μm; and C1 is 3.741772e. -16 C2 is 0.014388.
8. A computer program product, comprising a computer program, characterized in that: When executed by a processor, the program implements the steps of the multivariable fitting temperature calibration method for an uncooled long-wave infrared camera as described in any one of claims 1 to 7.
