An infrared wide-spectrum data calibration method and system using multi-temperature blackbody

Abstract

This invention discloses a method and system for calibrating infrared broadband data using multi-temperature blackbody, belonging to the field of optoelectronic information acquisition and processing. The invention utilizes a spectrum correlation detection device to collect data on targets at different temperatures. It was found that the system gain gradually decreases with changes in blackbody temperature, while the system bias remains essentially constant. Based on this, the gain and bias obtained from an ultra-low temperature blackbody are used to invert the spectrum of an ultra-low temperature target to obtain its radiance; the gain and bias obtained from a low temperature blackbody are used to invert the spectrum of a low temperature target to obtain its radiance; the gain and bias obtained from a medium temperature blackbody are used to invert the spectrum of a medium temperature target to obtain its radiance; and the gain and bias obtained from a high temperature blackbody are used to invert the spectrum of a high temperature target to obtain its radiance. Using different gains and biases to invert the true spectral characteristics (radiance) significantly improves the accuracy of subsequent identification algorithms.

Description

Technical Field

[0001] This invention belongs to the field of optoelectronic information acquisition and processing, and more specifically, relates to a method and system for calibrating infrared broadband data using a multi-temperature blackbody. Background Technology

[0002] Objects with different structures, compositions, and contents exhibit different physical and chemical properties, which in turn result in different spectral characteristics. Furthermore, different substances have different atomic and molecular structures, resulting in varying infrared photothermal effects and different selective absorption (radiation) characteristics, leading to significant differences in their light absorption and reflection. Therefore, spectroscopy is widely used in identifying and distinguishing different substances. This means that by measuring the infrared spectrum of an object and performing appropriate calculations and estimations, its relatively invariant spectral characteristics can effectively distinguish targets from the background and classify the target.

[0003] Zhang Tianxu's team proposed a nonlinear calibration method and device for infrared spectra of a spectral correlation system in patent CN111044153B, which acquires corresponding calibration data for ultra-low temperature targets, room temperature targets, medium temperature targets, and high temperature targets ranging from 1.7 to 14 μm.

[0004] However, this method uses blackbodies at different temperatures to calibrate targets at different temperatures, requiring prior calculation of the approximate temperature range of the target / object. Furthermore, using the same gain and bias to invert the target's radiance for targets at different temperatures does not accurately reflect the target's radiance, resulting in significant errors in the calculated radiance results. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a method and system for calibrating infrared broadband data using multi-temperature blackbody, aiming to solve the problem of large errors in existing calibration methods.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for calibrating infrared broadband data using a multi-temperature blackbody, comprising:

[0007] S1. Acquire infrared broadband data of blackbodies at different temperatures under different ambient temperatures, and collect multiple spectral data at each blackbodies temperature point;

[0008] S2. Denoise the spectrum containing abrupt changes;

[0009] S3. Using blackbody spectral data, calculate the system response gain and radiation bias at different blackbody temperatures under the same ambient temperature;

[0010] S4. Obtain the temperature of the detected part of the target object and determine the type of the target object as ultra-low temperature, low temperature, medium temperature, or high temperature; among them, when the temperature is in the range of [-20℃, 0℃), it is ultra-low temperature; when the temperature is in the range of [0℃, 30℃), it is low temperature; when the temperature is in the range of [30℃, 500℃), it is medium temperature; when the temperature is in the range of [500℃, 1000℃), it is high temperature.

[0011] S5. Using the gain and bias calculated from the blackbody in the corresponding temperature range, the target radiance of the corresponding type is obtained by inverting the target spectrum.

[0012] Preferably, step S2 is as follows:

[0013] S21. Determine the characteristic bands where abrupt changes occur based on the numerical value of DN;

[0014] S22. Denoise the characteristic bands using the following formula:

[0015]

[0016] Where λ is the feature band to be denoised, and DN λ ε is the digital signal value at the characteristic band λ, where ε is a positive integer.

[0017] Preferably, in step S3, the formula for calculating the system response gain is as follows:

[0018]

[0019] Where λ is the wavelength, T is the thermodynamic temperature in Kelvin, k(λ) is the spectral measurement system response gain at the target wavelength λ, and DN H S H These represent the digital signal value and radiance of a blackbody with a relatively high temperature, respectively; DN L S L These are the digital signal values ​​and radiance of a blackbody with a lower temperature, respectively.

[0020] Preferably, in step S3, the formula for calculating the system radiation bias is as follows:

[0021]

[0022] Where λ is the wavelength, T is the thermodynamic temperature in Kelvin, b(λ) is the radiation bias of the spectral measurement system at the target wavelength λ, and DN H S H These represent the digital signal value and radiance of a blackbody with a relatively high temperature, respectively; DN L S L These are the digital signal values ​​and radiance of a blackbody with a lower temperature, respectively.

[0023] Preferably, step S5 is as follows:

[0024] For ultra-low temperature targets, the mean DN values ​​of multiple samples of blackbodies at -20℃ and -10℃ are calculated respectively, and then the ultra-low temperature targets are radiatively corrected using the mean DN values ​​of these two blackbodies at different temperatures.

[0025] For low-temperature targets, the mean DN values ​​of multiple samples of blackbody at 30℃ and 80℃ are calculated respectively, and then the mean DN values ​​of these two blackbody at different temperatures are used to perform radiometric correction on the low-temperature targets.

[0026] For intermediate-temperature targets, the mean DN values ​​of multiple samples of blackbody at 300℃ and 500℃ are calculated respectively, and then the mean DN values ​​of these two blackbody at different temperatures are used to perform radiation correction on intermediate-temperature targets.

[0027] For high-temperature targets, the mean DN values ​​of multiple samples of blackbodies at 800℃ and 1000℃ are calculated respectively, and then the mean DN values ​​of these two blackbodies at different temperatures are used to perform radiation correction on the high-temperature targets.

[0028] Preferably, in step S5, the formula for calculating the image-side radiance of the target is as follows:

[0029]

[0030] Wherein, DN(λ) is the digital signal value measured by the spectral measurement system at the target wavelength λ, and k(λ) and b(λ) are the system response gain and system radiation bias at wavelength λ, respectively; S input (λ) represents the image-side radiance of a target with wavelength λ.

[0031] To achieve the above objectives, in a second aspect, the present invention provides an infrared broadband data calibration system utilizing a multi-temperature blackbody, comprising: a processor and a memory; the memory for storing computer execution instructions; and the processor for executing the computer execution instructions, causing the method described in the first aspect to be executed.

[0032] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0033] This invention proposes a calibration method and system for infrared broadband data using multi-temperature blackbody. By correlating the infrared broadband spectrum with detection equipment and finding targets at different temperatures, it was discovered that the system gain gradually decreases with changes in blackbody temperature, while the system bias remains essentially constant. Based on this, the gain and bias obtained using an ultra-low temperature blackbody are used to invert the spectrum of ultra-low temperature targets to obtain their radiance; the gain and bias obtained using a low temperature blackbody are used to invert the spectrum of low temperature targets to obtain their radiance; the gain and bias obtained using a medium temperature blackbody are used to invert the spectrum of medium temperature targets to obtain their radiance; and the gain and bias obtained using a high temperature blackbody are used to invert the spectrum of high temperature targets to obtain their radiance. Using different gains and biases to invert the true spectral characteristics (radiance) can significantly improve the accuracy of subsequent identification algorithms. Attached Figure Description

[0034] Figure 1 This is a flowchart of an infrared broadband data calibration method using a multi-temperature blackbody provided by the present invention.

[0035] Figure 2 These are the measured DN values ​​of the blackbody at 12 temperature points provided in this embodiment of the invention.

[0036] Figure 3 These are the DN values ​​of the blackbody at 12 temperature points after removing bad pixels, as provided in this embodiment of the invention.

[0037] Figure 4 This is the spectral energy conversion model of the infrared detection system provided in the embodiments of the present invention.

[0038] Figure 5 This is the DN value curve of a blackbody at an ambient temperature of 23℃ and an ambient temperature of -20℃ to 1000℃, provided in an embodiment of the present invention.

[0039] Figure 6 This is the radiance curve of a blackbody at an ambient temperature of 23℃ and an ambient temperature of -20℃ to 1000℃, provided in an embodiment of the present invention.

[0040] Figure 7 This is the system radiation response calculated using 30℃ and 80℃, as provided in the embodiments of the present invention.

[0041] Figure 8 This is the system radiation bias calculated using 30°C and 80°C, provided in an embodiment of the present invention.

[0042] Figure 9 This is the system radiation response calculated using 300℃ and 500℃, provided in the embodiments of the present invention.

[0043] Figure 10This is the system radiation bias calculated using 300℃ and 500℃, provided in the embodiments of the present invention.

[0044] Figure 11 This is the system radiation response calculated using 800℃ and 1000℃, as provided in the embodiments of the present invention.

[0045] Figure 12 This is the system radiation bias calculated using 800℃ and 1000℃, provided in the embodiments of the present invention.

[0046] Figure 13 The system gain curves obtained at different blackbody temperatures are provided in the embodiments of the present invention.

[0047] Figure 14 The system bias curves obtained from different blackbody temperatures are provided in the embodiments of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0049] like Figure 1 As shown, the present invention provides a method for calibrating infrared broadband data using a multi-temperature blackbody, comprising:

[0050] Step S1. Obtain infrared broadband data of blackbodies at different temperatures under different ambient temperatures, and collect multiple spectral data at each blackbody temperature point.

[0051] For N pairs of blackbody pairs, N sets of gains and biases of the system were obtained, and the nonlinear relationship between the N sets of gains and biases and the measured N pairs of blackbody pairs was analyzed. Furthermore, the N sets of gains and biases are also related to different ambient temperatures. Therefore, infrared broadband data of N sets of blackbody pairs at different ambient temperatures (T℃) were measured. M spectral data points were collected at each blackbody temperature point, with P sampling points in each spectrum. The blackbody data obtained in this study provides data support for subsequent calibration and inversion of targets at different temperatures.

[0052] Step S2. Denoise the spectrum with abrupt changes.

[0053] For P sampling points of the acquired blackbody spectrum, not every sampling point reflects the true spectral characteristics of the blackbody at that temperature; instead, it may become a characteristic of the blackbody. Therefore, it is necessary to denoise some of the P sampling points to represent the true characteristics of the blackbody at that temperature. The denoising formula is as follows:

[0054]

[0055] Where λ is the feature band to be denoised, and DN λ ε is the digital signal value at the characteristic band λ, where ε is a positive integer.

[0056] Step S3. Using blackbody spectral data, calculate the system response gain and radiation bias for different blackbody temperatures under the same ambient temperature.

[0057] At an ambient temperature of T℃, for a blackbody at N temperature points, using the blackbody spectral data at different temperatures, a total of Q = N / 2 different system gains and biases are calculated. The relevant system parameters, system response gain, and system radiation bias are obtained from the blackbody signal values ​​measured during calibration, as shown in the following formulas:

[0058]

[0059]

[0060] In the above formula, DN H S H These represent the digital signal value and radiance of a high-temperature blackbody, respectively; DN L S L These represent the digital signal values ​​and radiance of a lower-temperature blackbody, respectively. A total of Q sets of different system gains and biases are calculated, providing important parameters for subsequent inversion of ultra-low temperature targets, normal temperature targets, medium temperature targets, and high temperature targets.

[0061] Step S4. Obtain the temperature of the detected part of the target object and determine whether the target object is of ultra-low temperature, low temperature, medium temperature or high temperature.

[0062] Temperatures ranging from -20℃ to 0℃ are considered extremely low; temperatures ranging from 0℃ to 30℃ are considered low; temperatures ranging from 30℃ to 500℃ are considered moderate; and temperatures ranging from 500℃ to 1000℃ are considered high.

[0063] Step S5. Using the gain and bias calculated by the blackbody in the corresponding temperature range, the target spectrum of the corresponding type is inverted to obtain the target radiance.

[0064] The gain and bias obtained using an ultra-low temperature blackbody are used to invert the spectrum of an ultra-low temperature target to obtain its radiance. The gain and bias obtained using a low temperature blackbody are used to invert the spectrum of a low temperature target to obtain its radiance. The gain and bias obtained using a medium temperature blackbody are used to invert the spectrum of a medium temperature target to obtain its radiance. The gain and bias obtained using a high temperature blackbody are used to invert the spectrum of a high temperature target to obtain its radiance. The inversion formulas are as follows:

[0065] DN(λ)=k(λ)·S input (λ)+b(λ)

[0066] In the above formula, DN(λ) is the digital signal value measured by the spectral measurement system at the target wavelength λ, and k(λ) and b(λ) are the system response gain and system radiation bias at wavelength λ, respectively; S input (λ) represents the image-side radiance of the target.

[0067] Example

[0068] 1. Spectral measurement of a broad-spectrum blackbody

[0069] Spectral data were collected at 12 blackbody temperature points (-20℃, -10℃, 0℃, 5℃, 30℃, 50℃, 80℃, 100℃, 300℃, 500℃, 800℃, and 1000℃) under three different ambient temperatures: -1℃, 23℃, and 27℃. More than 600 spectra were collected for each blackbody temperature point. This calibration experiment provides important support for subsequent analysis. The measured DN values ​​of the blackbody at the 12 temperature points are as follows: Figure 2 As shown.

[0070] 2. Denoising processing in characteristic bands

[0071] Based on the measured data, it can be seen that in this embodiment, regardless of the blackbody temperature change, there is always a bad spot at 2.52 μm. Therefore, the bad spot near λ = 2.52 μm is smoothed, and the DN value near wavelength λ = 2.52 μm is trend-fitted with the DN values ​​near 2.52 μm and 2.53 μm. The DN values ​​of the blackbody at the 12 temperature points after removing the bad spot are as follows: Figure 3 As shown.

[0072] 3. Calculate the system gain and bias at different temperatures.

[0073] The spectral energy conversion relationship of the infrared detection system is as follows: Figure 4 As shown, light energy, after passing through an optical system, a photoelectric conversion system, and numerical quantization, yields the sensor's spectral response DN value, which physically represents a voltage or current signal. The intensity of target reflection / radiation is qualitatively reflected by its value. Different detection systems respond differently to the same light energy, which is related to the system's own response characteristics. Radiometric calibration establishes a quantitative relationship between target reflection / radiation energy and system response using standard blackbody spectral data. This relationship is then used to perform radiometric correction on the measured spectra of different types of targets, inverting the system response DN value to the target's radiation intensity, facilitating quantitative research on system spectral data.

[0074] The two-point calibration method is based on the linear relationship between the system response DN value and the target radiated / reflected energy over the dynamic range. This linear relationship is then determined using blackbody spectral data to obtain the parameters for radiometric calibration. The system radiance conversion is shown in the following equation:

[0075] DN(λ)=k(λ)·L(λ)+b(λ)

[0076] In the above formula, DN(λ) is the digital signal value measured by the spectral measurement system at the target wavelength λ, k(λ) and b(λ) are the system response gain and system radiation bias at the wavelength λ, respectively; L(λ) is the true radiance of the target.

[0077] By using the blackbody signal value measured during calibration, the relevant parameters of the system, the system response gain, and the system radiation bias are obtained, as shown in the following formulas:

[0078]

[0079]

[0080] In the above formula, DN H (λ), L H (λ,T) represent the digital signal value and radiance of the high-temperature blackbody, respectively. D (λ), L D (λ,T) represent the digital signal value and radiance of the low-temperature blackbody, respectively.

[0081] The advantage of this method is its low computational cost. However, a disadvantage is that when the temperature of the measured blackbody is unstable, a small measurement error can lead to a large error after blackbody calibration. The radiance can be calculated using Planck's formula, as follows:

[0082]

[0083] Where L(λ,T) is the blackbody radiance in W / m. 2 ·um, λ is the wavelength (um), T is the temperature of the blackbody (K), c1 and c2 are Planck's constants, c1 = 3.7418 × 10 8 (W / m 2 ·um4), c2=1.4387752×10 4 (K·um).

[0084] Based on the above formula, the spectral DN value curves and radiance curves of blackbodies at different temperatures under the same ambient temperature can be obtained, as shown below. Figure 5 , Figure 6 As shown.

[0085] Data fitting is an important data processing method, with polynomial curve fitting being the most commonly used. However, when there are many data points, a low polynomial order results in unsatisfactory fitting accuracy and performance. Improving fitting accuracy and performance requires increasing the curve order, but a high order introduces computational complexity and other disadvantages. Therefore, using only one polynomial curve function to fit a large amount of data is insufficient to achieve good fitting accuracy and performance. To effectively address this issue, piecewise curve fitting is generally employed. Previous piecewise curve fitting methods were primarily used for data measured in the natural sciences, where variations generally follow certain patterns. Therefore, when fitting these measurement data, traditional piecewise curve fitting methods typically involve first analyzing the data based on subjective experience and then performing the fitting. However, for data from some practical problems, the mechanisms of these variations are often very complex and do not exhibit the strict regularity of physical laws, resulting in significant uncertainty. Based on the calibration data obtained from the aforementioned radiometric calibration experiment, and combining the radiative transfer model and conversion model of the remote sensing system, a radiometric calibration algorithm is used to complete the calibration task. The main purpose of radiometric calibration algorithms is to fit the measured calibration data with the input conditions and calculate the calibration coefficients. Radiometric calibration algorithms are generally divided into statistical methods and computational methods. Statistical methods require a large amount of calibration data, are computationally complex, have poor timeliness, and their calibration accuracy is affected by many factors in the calibration experiment, so they are rarely used in engineering.

[0086] For low-temperature targets, the average spectra of 30℃ and 80℃ blackbodies are used to perform radiometric correction on the low-temperature targets. For medium-temperature targets, the average spectra of 300℃ and 500℃ blackbodies are used to perform radiometric correction on the measured spectral data. For high-temperature targets, the average spectra of 800℃ and 1000℃ blackbodies are used to perform radiometric correction on the measured spectral data.

[0087] The DN values ​​of a 30℃ and 80℃ blackbody were collected and averaged. Then, the average value of these two blackbody values ​​was used to correct for low-temperature conditions. The resulting system radiation response and system radiation bias were fitted using a 15th-order term. The resulting graphs are shown below. Figure 7 , Figure 8 As shown.

[0088] The average values ​​of the blackbody DN at 300℃ and 500℃ were obtained and then used to correct the intermediate-temperature target. The resulting system radiation response and system radiation bias are shown below. Figure 9 , Figure 10 As shown.

[0089] The DN values ​​of the collected 800℃ and 1000℃ blackbody were averaged, and then the average value of these two blackbody values ​​was used to correct the high-temperature target. The resulting system radiation response and system radiation bias are shown below. Figure 11 , Figure 12 As shown.

[0090] Analyze the gains and biases obtained above, such as... Figure 13 and 14 As shown, the system gain decreases as the blackbody temperature increases, while the system bias remains essentially unchanged with the blackbody temperature.

[0091] 4. Select different gains and biases to invert and calibrate targets at different temperatures.

[0092] The average DN values ​​of the collected -20℃ and -10℃ blackbodies are calculated separately, and then the average value of these two blackbodies at different temperatures is used to correct the ultra-low temperature target. The average DN values ​​of the collected 30℃ and 80℃ blackbodies are calculated separately, and then the average value of these two blackbodies at different temperatures is used to correct the low temperature target. The average DN values ​​of the collected 300℃ and 500℃ blackbodies are calculated separately, and then the average value of these two blackbodies at different temperatures is used to correct the medium temperature target. The average DN values ​​of the collected 800℃ and 1000℃ blackbodies are calculated separately, and then the average value of these two blackbodies at different temperatures is used to correct the high temperature target.

[0093] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An infrared wide spectral data calibration method using a multi-temperature blackbody, characterized by, include: S1. Acquire infrared broadband data of blackbodies at different temperatures under different ambient temperatures, and collect multiple spectral data at each blackbodies temperature point; S2. Denoise the spectrum containing abrupt changes; S3. Using blackbody spectral data, calculate the system response gain and radiation bias at different blackbody temperatures under the same ambient temperature; S4. Obtain the temperature of the detected site of the target object, and determine the type of the target object as ultra-low temperature, low temperature, medium temperature or high temperature; wherein the temperature belongs to ultra-low temperature; the temperature belongs to low temperature; the temperature belongs to medium temperature; and the temperature belongs to high temperature. S5. Using the gain and bias calculated from the blackbody in the corresponding temperature range, the target radiance of the corresponding type is obtained by inverting the target spectrum.

2. The method as described in claim 1, characterized in that, Step S2 is as follows: S21. Determine the characteristic bands where abrupt changes occur based on the numerical value of DN; S22. Denoise the characteristic bands using the following formula: in, For the characteristic bands to be denoised, Characteristic bands The digital signal value of the blackbody. It is a positive integer.

3. The method as described in claim 1, characterized in that, In step S3, the formula for calculating the system response gain is as follows: in, For wavelength, Temperature is the thermodynamic temperature, and the unit is Kelvin. For the target wavelength is The system response gain at the measurement spectrum, , These are the digital signal values ​​and radiance of a blackbody with a relatively high temperature, respectively. , These are the digital signal values ​​and radiance of a blackbody with a lower temperature, respectively.

4. The method as described in claim 1, characterized in that, In step S3, the formula for calculating the system radiation bias is as follows: in, For wavelength, Temperature is the thermodynamic temperature, and the unit is Kelvin. For the target wavelength is Radiation bias of the spectral measurement system , These are the digital signal values ​​and radiance of a blackbody with a relatively high temperature, respectively. , These are the digital signal values ​​and radiance of a blackbody with a lower temperature, respectively.

5. The method as described in claim 2, characterized in that, Step S5 is as follows: For ultra-low temperature targets, the mean DN values ​​of multiple samples of blackbodies at -20℃ and -10℃ are calculated respectively, and then the ultra-low temperature targets are radiatively corrected using the mean DN values ​​of these two blackbodies at different temperatures. For low-temperature targets, the mean DN values ​​of multiple samples of blackbody at 30℃ and 80℃ are calculated respectively, and then the mean DN values ​​of these two blackbody at different temperatures are used to perform radiometric correction on the low-temperature targets. For intermediate-temperature targets, the mean DN values ​​of multiple samples of blackbody at 300℃ and 500℃ are calculated respectively, and then the mean DN values ​​of these two blackbody at different temperatures are used to perform radiation correction on intermediate-temperature targets. For high-temperature targets, the mean DN values ​​of multiple samples of blackbodies at 800℃ and 1000℃ are calculated respectively, and then the mean DN values ​​of these two blackbodies at different temperatures are used to perform radiation correction on the high-temperature targets.

6. The method as described in claim 1, characterized in that, In step S5, the formula for calculating the image-side radiance of the target is as follows: in, For the target wavelength is The digital signal value of the blackbody measured by the spectral analysis system. , The wavelengths are respectively The system response gain and system radiation bias at the location; For wavelength Image-side radiance of the target.

7. A calibration system for infrared broadband data utilizing a multi-temperature blackbody, characterized in that, include: Processor and memory; The memory is used to store computer-executed instructions; The processor is configured to execute the computer execution instructions, causing the method described in any one of claims 1 to 6 to be executed.

Images