A method for field radiometric calibration of an infrared imager

Abstract

This invention discloses an outdoor radiometric calibration method for an infrared imager, comprising: under outdoor operating conditions, selecting observation times that meet preset radiometric calibration conditions as radiometric calibration times; within the radiometric calibration times, based on the near-time-series acquisition characteristics of sky observation signals and outer blackbody observation signals, performing differential processing on the two to obtain measured signal difference components; based on the outdoor environmental information corresponding to the radiometric calibration times, constructing a regional clear-sky atmospheric model, and using a radiative transfer model to simulate and obtain the downward infrared radiation of the clear-sky all-sky at the corresponding times, and constructing physical radiation difference components for outdoor radiometric calibration; based on multiple sets of measured signal difference components and corresponding physical radiation difference components, obtaining the full-field radiometric calibration coefficient matrix through regression analysis, and converting the infrared imaging signal into infrared radiation. This method eliminates the need to construct complex laboratory radiometric calibration conditions, allowing for direct radiometric calibration of the infrared imager in the outdoor environment.

Description

Technical Field

[0001] This invention relates to the field of infrared remote sensing and quantitative radiation measurement technology, and in particular to an external field radiation calibration method for an infrared imager. Background Technology

[0002] In the 8-14 μm infrared atmospheric window band, infrared imagers can observe and distinguish the radiation characteristics of the sky and clouds, making them an important observational tool for applications such as cloud cover, cloud height, cloud optical thickness (LOC), and sky brightness temperature monitoring. These applications rely not only on the spatial distribution characteristics of the infrared imaging results but also on whether the radiation quantity corresponding to the observed signal has a clear and reliable physical meaning. In practical applications, statistical analysis based solely on the digital quantization (DN) values ​​output by the infrared imager, such as distinguishing between clear skies and cloud areas using empirical thresholds, can achieve cloud cover discrimination under certain conditions. However, this method is essentially image processing and struggles to maintain consistency across different regions, seasons, or atmospheric conditions. It also cannot support quantitative inversion of cloud height, cloud optical thickness, and other parameters dependent on radiative transfer processes. In contrast, converting infrared imaging signals into physically meaningful radiation quantities and analyzing them in conjunction with regional climate characteristics or radiative transfer models allows for the construction of a closed-loop physical framework. This is a more reasonable and reliable technical approach for conducting cloud parameter observation and inversion and sky brightness temperature monitoring based on this band. Therefore, accurate and stable radiometric calibration of infrared imagers is a fundamental prerequisite for realizing the above-mentioned quantitative applications.

[0003] Radiometric calibration of infrared imagers is typically performed under laboratory conditions. The basic principle involves the imager conducting multiple observations of a standard blackbody source at different temperatures to establish a correspondence between the imaging signal and the ideal blackbody radiation, thereby obtaining radiometric calibration coefficients. This method can achieve high-precision radiometric calibration results under controlled conditions. However, laboratory calibration usually only covers a limited sensor operating temperature range. During long-term operation in actual outdoor environments, especially in cold regions or environments with large diurnal temperature variations, the temperature of the infrared imager's detector focal plane array changes significantly, rendering the radiometric calibration parameters obtained in the laboratory inapplicable under outdoor conditions, thus introducing serious systematic biases into the radiation values. Although multi-temperature calibration conditions can be constructed in the laboratory to extend the calibration of the infrared imager across different sensor temperature ranges, this approach often requires an additional fully enclosed temperature-controlled chamber and a more complex calibration process, significantly increasing the construction cost and implementation difficulty of the radiometric calibration system. Summary of the Invention

[0004] In view of the aforementioned problems in the radiometric calibration of existing infrared imagers, the purpose of this invention is to provide an external field radiometric calibration method for infrared imagers.

[0005] To achieve the aforementioned objectives, the technical solution adopted by this invention is: an external field radiation calibration method for an infrared imager, comprising: Step 1: Under field operating conditions, screen the historical observation times formed by the infrared imager and select the observation times that meet the preset radiometric calibration conditions as radiometric calibration times; the preset radiometric calibration conditions include at least clear night sky, light wind or no wind, good visibility and low humidity conditions. Step 2: During the radiometric calibration period, based on the characteristic that the sky observation signal and the external blackbody observation signal are acquired in pairs within a short time interval, the two are differentially processed to obtain the measured signal difference component for field radiometric calibration. Step 3: Based on the external environmental information corresponding to the radiometric calibration time, construct a regional clear sky atmospheric model, and use the radiative transfer model to simulate and obtain the clear sky down-infrared radiation of the corresponding time. At the same time, calculate the blackbody radiation based on the blackbody temperature, and construct the physical radiation difference component for external radiometric calibration based on the radiation difference between the clear sky down-infrared radiation and the blackbody radiation. Step 4: Based on the multiple sets of measured signal difference components obtained in Step 2 and the corresponding physical radiation difference components constructed in Step 3, the full-field radiation calibration coefficient matrix of the infrared imager is obtained through regression analysis, and a quantitative mapping relationship between infrared imaging signals and physical radiation quantities is established, converting the infrared imaging signals into physical radiation quantities.

[0006] Furthermore, in step 1, the low humidity condition is determined by estimating the total water vapor content of the atmospheric column corresponding to the radiation calibration time. Only when the total water vapor content of the atmospheric column does not exceed the preset threshold is the radiation calibration time selected for external field radiation.

[0007] Furthermore, in step 3, when constructing the regional clear-sky atmospheric model, the regional clear-sky atmospheric model is constrained based on the ground meteorological observation data corresponding to the radiation calibration time and the local or regional atmospheric profile information, so that the regional clear-sky atmospheric model matches the radiation calibration time.

[0008] Furthermore, in step 4, the radiation calibration coefficients are linear calibration coefficients that convert infrared imaging signals into radiation quantities. The regression analysis obtains the linear calibration coefficients based on the correspondence between multiple sets of paired observation data and the corresponding simulated radiation and outer blackbody radiation under clear sky.

[0009] Furthermore, the method also includes: Step 5: constraining the applicable range of the radiometric calibration coefficients so that the radiometric calibration results are only output and applied within the preset observation zenith angle range.

[0010] Compared with the prior art, the present invention has the following beneficial effects: (1) Based on the pairwise and near-time-series acquisition characteristics of sky observation signal and external blackbody observation signal, the present invention constructs the measured signal difference component and combines it with the corresponding physical radiation difference component for regression analysis, which can effectively eliminate the influence of internal noise of infrared imaging module on radiation calibration results and improve the stability and reliability of external field radiation calibration.

[0011] (2) This invention applies joint constraints such as clear night sky, low humidity, and good visibility to the calibration observation time, and uses the radiation transfer model to simulate the downward infrared radiation of clear sky under environmental information constraints, thereby reducing the impact of atmospheric water vapor and aerosol uncertainty on the radiation calibration results and making the field radiation calibration process have clear physical consistency.

[0012] (3) This invention does not require the construction of complex laboratory radiation calibration conditions. It can directly complete the radiation calibration of infrared imagers or the correction of laboratory radiation calibration coefficients based on field operation data, and stably convert infrared imaging signals into radiation quantities with physical meaning. This provides a reliable quantitative basis for applications such as cloud parameter inversion and sky brightness temperature monitoring based on radiation quantities, and reduces the cost and engineering complexity of calibration implementation. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0014] Figure 1 This is a flowchart illustrating an external field radiation calibration method for an infrared imager according to the present invention. Figure 2 This is a schematic diagram of the working state of the infrared imager of the present invention in a single time period; Figure 3 This is a schematic diagram illustrating the effect of the differential signal suppression imaging module of the present invention on the internal noise. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of protection.

[0016] Please see Figure 1 This invention discloses an external field radiation calibration method for an infrared imager, comprising: Step 1: Under field operating conditions, screen the historical observation times formed by the infrared imager and select the observation times that meet the preset radiometric calibration conditions as radiometric calibration times; the preset radiometric calibration conditions include at least clear night sky, light wind or no wind, good visibility and low humidity conditions.

[0017] Specifically, this embodiment applies to infrared imagers operating in the 8-14 μm band, equipped with an external blackbody calibration structure, and whose full field-of-view angle calibration has been completed. Before performing external field radiation calibration, the observation times used for calibration are screened and constrained. First, clear sky determination can be based on visible light imaging data acquired synchronously with the infrared imager observations, artificial meteorological observation records, or a comprehensive judgment based on the spatial distribution characteristics of sky radiation in infrared images. Although in the 8-14 μm infrared atmospheric window band, the influence of the sun on the sky background radiation under daytime conditions usually manifests as bright spots with clear boundaries and limited spatial range, which can be processed by removing affected pixels and reconstructing by interpolation, this embodiment selects nighttime observation times to avoid introducing additional artificial correction operations into the original observed DN data during the radiometric calibration process. At the same time, to reduce the influence of wind-induced convection on the surface temperature distribution of the outer blackbody, it is preferable to select observation times under windless or weak wind conditions to improve the temperature uniformity and stability of the outer blackbody during the calibration observation process. In addition, based on the sensitivity test results of the radiative transfer mode, the influence of aerosol changes on the downward infrared radiation under clear sky in the 8-14 μm band is relatively limited, and only in extreme aerosol events with extremely low horizontal visibility will it produce perceptible deviations in the clear sky radiation simulation results. In this embodiment, observation times with good visibility are selected for field radiometric calibration to effectively eliminate the influence of aerosol factors on the clear sky radiation simulation and radiometric calibration results. The visibility information can be obtained from the observation results of the horizontal visibility meter or artificial meteorological observation records.

[0018] In this embodiment, low humidity conditions are determined by estimating the total water vapor content of the atmospheric column corresponding to the radiation calibration time. Only when the total water vapor content of the atmospheric column does not exceed a preset threshold is the time selected for field radiation calibration, so as to constrain the sensitivity of clear sky radiation simulation to water vapor uncertainty.

[0019] Specifically, given the significant impact of atmospheric water vapor content on clear-sky downward infrared radiation, atmospheric water vapor conditions are a key constraint. The total atmospheric water vapor content is estimated based on the monthly average water vapor profile obtained from climate statistics or reanalysis data, combined with measured surface temperature and humidity data corresponding to the observation time. In this embodiment, the estimation process of the total atmospheric water vapor content specifically includes: obtaining the monthly average water vapor profile characteristics of the area where the observation station is located based on ERA5 reanalysis data, which is used to characterize the typical vertical distribution pattern of water vapor in the area, and determining the characteristic scale of water vapor attenuation with altitude accordingly; simultaneously, using the surface temperature, air pressure, and relative humidity data obtained synchronously during the observation time, the surface water vapor density is calculated. The surface water vapor density is combined with the water vapor vertical characteristic scale determined based on climatological data to obtain the estimated value of the total atmospheric water vapor content corresponding to the observation time. When available measured water vapor vertical profile data (such as radiosonde observations) exists, the water vapor profile can also be directly vertically integrated to obtain the total atmospheric water vapor content. This invention does not limit the method of obtaining water vapor content.

[0020] Simulation results of atmospheric downward infrared radiative transfer at 8-14 μm indicate that when the total water vapor content in the atmospheric column is low, even if there is a certain deviation between the atmospheric water vapor content obtained based on the estimation method and the actual situation, its impact on the simulation results of clear-sky downward infrared radiation remains limited. On the one hand, under low water vapor background conditions, the absolute deviation of atmospheric water vapor estimation is usually small; on the other hand, the contribution of water vapor to clear-sky downward infrared radiation is itself at a low level, making it difficult to amplify the clear-sky radiation simulation deviation introduced by water vapor estimation errors. This helps to ensure the stability and reliability of the field radiation calibration results. In this embodiment, observation times with a total atmospheric column water vapor content of less than 1 g / cm² were selected for subsequent field radiation calibration.

[0021] Step 2: During the radiometric calibration period, based on the near-time acquisition characteristics of the sky observation signal and the external blackbody observation signal, differential processing is performed on the two to obtain the measured signal difference component for external radiometric calibration.

[0022] Please see Figure 2 In this embodiment, the infrared imager switches between two shooting states according to a preset operating mode within each observation period. Specifically, within the same observation period, the infrared imager first takes a picture of the blackbody while it is blocking the observation optical path, acquiring the corresponding blackbody observation signal; then, after the blackbody moves out of the observation optical path, it takes a picture of the sky, acquiring the corresponding sky observation signal. The aforementioned blackbody and sky images are completed by the same imaging module, using the same imaging parameters and data acquisition methods. The paired, near-time-series acquisition of blackbody-blocked observation data and sky observation data is used to eliminate or reduce the influence of internal noise of the infrared imager on the radiometric calibration results.

[0023] The DN (Digital Quantization) value acquired by the infrared imaging module is the digital response of the infrared imager to incident infrared radiation. For any pixel (i, j) in the imaging module, its DN value can be expressed as the superposition of the target radiation term, the thermal radiation interference term within the imaging system, and the non-uniform noise term of the infrared array pixels. Specifically, the DN value can be expressed as: Where L(λ,i,j) is the target spectral radiance in the pixel direction, R(λ) is the detector's spectral response function, G(i,j) is the pixel gain, B is the pixel bias term, and N... NU (i,j) represents the non-uniform noise term of the pixel response; L sys (λ) is a system radiation interference term related to the internal state of the infrared imaging module, used to characterize the additional thermal radiation contribution introduced by the optical system, cavity structure and detector itself.

[0024] Please see Figure 3 , Figure 3 (a) is the external blackbody DN signal. Figure 3 (b) is the sky DN signal. Figure 3 (c) is the DN differential signal. In this embodiment, the external blackbody occlusion shooting and the sky shooting are acquired in close time sequence. The two shootings are completed within about 10 seconds. The thermal noise and pixel response non-uniform noise inside the imaging module can be approximately regarded as unchanged during the two shooting processes. Therefore, they can be effectively eliminated and weakened in differential processing, so that the differential result mainly reflects the response difference between sky radiation and external blackbody radiation.

[0025] Step 3: Based on the external environmental information corresponding to the radiometric calibration time, construct a regional clear sky atmospheric model, and use the radiative transfer model to simulate and obtain the clear sky down-infrared radiation of the corresponding time. At the same time, calculate the blackbody radiation based on the blackbody temperature, and construct the physical radiation difference component for external radiometric calibration based on the radiation difference between the clear sky down-infrared radiation and the blackbody radiation.

[0026] Furthermore, when constructing a regional clear-sky atmospheric model, the regional clear-sky atmospheric model is constrained based on the ground meteorological observation data corresponding to the radiation calibration time and the local or regional atmospheric profile information, so that the regional clear-sky atmospheric model matches the radiation calibration time.

[0027] In this embodiment, the simulation of clear-sky down-heaven infrared radiation is implemented using the libRadtran radiative transfer mode. The regional clear-sky atmospheric model used for simulation is generated based on ERA5 reanalysis data. Specifically, it constructs a monthly average atmospheric profile with climatological significance using the monthly average atmospheric elements of the observation station's region to characterize the typical atmospheric structure of the region in the corresponding month. Based on this, combined with surface meteorological observation data acquired synchronously during radiometric calibration, the total water vapor content of the atmospheric column is estimated, and the estimated water vapor information is used as input parameters to modulate the water vapor conditions of the monthly average atmospheric model. Through the radiative transfer mode, under preset atmospheric conditions and water vapor content constraints, the distribution characteristics of clear-sky down-heaven infrared radiation with zenith angle variation at the corresponding radiometric calibration time are simulated. Furthermore, utilizing the full-field-of-view calibration information already completed by the infrared imager, the clear-sky infrared radiation simulation curve is interpolated and mapped according to the zenith angle corresponding to each pixel, thereby obtaining a clear-sky infrared radiation simulation result that is consistent with the instrument's field of view geometry and covers the entire imaging field of view. Simultaneously, the outer blackbody radiation value is directly calculated based on the measured temperature of the sub-outer blackbody during radiation calibration. This calculation is based on an ideal blackbody radiation model and, combined with the infrared imager's field of view and observation geometry, yields an outer blackbody radiation result consistent with the imager's field of view. Based on this, a physical radiation difference component is constructed for external field radiation calibration, used for subsequent regression analysis and solution of radiation calibration coefficients.

[0028] Step 4: Based on the multiple sets of measured signal difference components obtained in Step 2 and the corresponding physical radiation difference components constructed in Step 3, the full-field radiation calibration coefficient matrix of the infrared imager is obtained through regression analysis, and a quantitative mapping relationship between infrared imaging signals and physical radiation quantities is established, converting the infrared imaging signals into infrared radiation quantities.

[0029] In this embodiment, the radiometric calibration coefficients are linear calibration coefficients that convert infrared imaging signals into radiance. Regression analysis is used to obtain these linear calibration coefficients based on the correspondence between multiple sets of paired observation data and the corresponding simulated all-sky downward radiation and outer blackbody radiation. Specifically, based on multiple sets of measured signal difference components and their corresponding physical radiation difference components, regression analysis is used to obtain the full-field radiometric calibration coefficient matrix of the infrared imager. The regression relationship is as follows: in, k i,j , b i,j These are the scaling factors. This represents the difference component of the measured signal; This is the physical radiation difference component; The measured signal difference components are converted into sky-down infrared radiation using the full-field radiometric calibration coefficient matrix. Specifically, for pixel (i, j), the observed measured signal difference components are substituted into the calibration coefficients obtained from regression. k i,j and b i,j Inversion calculations are performed to obtain the downward infrared radiance of the sky for this pixel, and its expression is: Among them, L i,j sky ΔDN represents the downward infrared radiance of the sky in the direction of the (i, j)th pixel. i,j obs The measured digital quantization difference between the sky observation signal and the outer blackbody observation signal at pixel (i, j) represents the difference between the two signals; h is Planck's constant; c is the speed of light in vacuum; k B T is the Boltzmann constant; ext The measured temperature of the outer blackbody is given; the integral term represents the ideal blackbody radiance calculated from the outer blackbody temperature in the 8-14 μm band.

[0030] Step 5: Constrain the applicable range of the radiometric calibration coefficients so that the radiometric calibration results can only be output and applied within the preset observation zenith angle range.

[0031] Specifically, existing radiative transfer simulation capability assessment results show that within a small to medium zenith angle range, the simulation results of clear-sky downward infrared radiation are generally stable and can reflect the true atmospheric radiation characteristics well. However, as the zenith angle increases further, the cumulative error and mode uncertainty in the radiative transfer process are significantly amplified due to the rapid increase in the optical path. Although the difference component of the measured signal between the outer blackbody and the sky is stable in the imager's field of view edge region, and the outer blackbody radiation calculation itself is not affected by the zenith angle, the systematic error introduced by the clear-sky radiation simulation deviation may reduce the reliability of the radiometric calibration results in the edge field of view region. In this embodiment, the zenith angle threshold is set to 75°, corresponding to an effective field of view angle of approximately 150°, which covers the vast majority of all-sky observation areas and can meet the actual needs of cloud-related business applications. For infrared imagers or non-wide-angle infrared imaging systems with a small field of view angle and a maximum zenith angle lower than the threshold, their entire effective field of view is within the reliable radiation simulation range, and therefore, no additional applicable range constraints are required.

[0032] The present invention has the following beneficial effects: (1) Based on the pairwise near-time acquisition characteristics of sky observation signal and external blackbody observation signal, the present invention constructs the measured signal difference component and combines it with the corresponding physical radiation difference component for regression analysis, which can effectively eliminate the influence of internal noise of infrared imaging module on radiation calibration results and improve the stability and reliability of external field radiation calibration.

[0033] (2) This invention applies joint constraints such as clear night sky, low humidity, and good visibility to the calibration observation time, and uses the radiation transfer model to simulate the downward infrared radiation of clear sky under environmental information constraints, thereby reducing the impact of atmospheric water vapor and aerosol uncertainty on the radiation calibration results and making the field radiation calibration process have clear physical consistency.

[0034] (3) This invention does not require the construction of complex laboratory radiation calibration conditions. It can directly complete the radiation calibration of infrared imagers or the correction of laboratory radiation calibration coefficients based on field operation data, and stably convert infrared imaging signals into radiation quantities with physical meaning, i.e. infrared radiation quantities. This provides a reliable quantitative basis for applications such as cloud parameter inversion and sky brightness temperature monitoring based on radiation quantities, and reduces the calibration implementation cost and engineering complexity.

[0035] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort, such as modifications to the technical solutions described in the following embodiments or equivalent substitutions of some technical features, are within the scope of protection of the present invention.

Claims

1. A method for external field radiometric calibration of an infrared imager, characterized in that, include: Step 1: Under field operating conditions, screen the historical observation times generated by the infrared imager and select the observation times that meet the preset radiometric calibration conditions as the radiometric calibration times; Step 2: During the radiometric calibration period, based on the pairing and near-time-series acquisition characteristics of the sky observation signal and the external blackbody observation signal, differential processing is performed on the two to obtain the measured signal difference component for field radiometric calibration. Step 3: Based on the external environmental information corresponding to the radiometric calibration time, construct a regional clear sky atmospheric model, and use the radiative transfer model to simulate and obtain the clear sky down-infrared radiation of the corresponding time. At the same time, calculate the blackbody radiation based on the blackbody temperature, and construct the physical radiation difference component for external radiometric calibration based on the radiation difference between the clear sky down-infrared radiation and the blackbody radiation. Step 4: Based on the multiple sets of measured signal difference components obtained in Step 2 and the corresponding physical radiation difference components constructed in Step 3, the full-field radiation calibration coefficient matrix of the infrared imager is obtained through regression analysis, and a quantitative mapping relationship between infrared imaging signals and physical radiation quantities is established, converting the infrared imaging signals into infrared radiation quantities.

2. The method for external field radiation calibration of an infrared imager according to claim 1, characterized in that, The preset radiation calibration conditions in step 1 include at least clear skies at night, light or no wind, good visibility, and low humidity.

3. The method for external field radiation calibration of an infrared imager according to claim 1, characterized in that, In step 1, the low humidity condition is determined by estimating the total water vapor content of the atmospheric column corresponding to the radiation calibration time. Only when the total water vapor content of the atmospheric column does not exceed the preset threshold is the radiation calibration time selected for external field radiation.

4. The method for external field radiation calibration of an infrared imager according to claim 1, characterized in that, In step 3, when constructing the regional clear-sky atmospheric model, the clear-sky atmospheric model is constrained based on the ground meteorological observation data corresponding to the radiation calibration time and the local or regional atmospheric profile information, so that the regional clear-sky atmospheric model matches the radiation calibration time.

5. The method for external field radiation calibration of an infrared imager according to claim 1, characterized in that, In step 4, the radiation calibration coefficients are linear calibration coefficients that convert infrared imaging signals into radiation quantities. The regression analysis obtains the linear calibration coefficients based on the correspondence between multiple sets of paired observation data and the corresponding simulated radiation and outer blackbody radiation under clear sky.

6. The method for external field radiation calibration of an infrared imager according to claim 1, characterized in that, The method further includes: Step 5: constraining the applicable range of the radiometric calibration coefficients so that the radiometric calibration results are only output and applied within the preset observation zenith angle range.

Images