Gas early warning and radiation calibration system and method based on uncooled infrared camera

Abstract

The application provides a gas early warning and radiation calibration system and method based on a non-cooled infrared camera, and the system comprises a non-cooled infrared camera, a band-pass filter, a light-shield shutter, a lens, and through camera non-uniformity correction, instrument internal self-radiation correction, voltage value-temperature value conversion, temperature data of different wave bands are acquired to perform system radiation calibration; an 8-14 mu m infrared atmospheric window wave band is adopted, based on passive infrared telemetry principle, taking a natural object as a background, without additional light sources; the system adopts multiple wide-band filters to acquire multispectral information of the gas, and according to the multispectral information, target component identification and early warning are performed. The application solves the technical problems that the existing technology has poor infrared calibration effect, leads to low gas imaging telemetry precision, and needs to use an active light source.

Description

Technical Field

[0001] This invention relates to the field of gas imaging telemetry and detection, specifically to a gas early warning and radiation calibration system and method based on an uncooled infrared camera. Background Technology

[0002] Leaks of hazardous chemicals pose a serious threat to people's lives and property. Monitoring and early warning of hazardous chemicals are of great importance in industrial production, emergency response, rescue, and protection.

[0003] While existing online monitoring technologies for hazardous gases offer high sensitivity, on-site sampling requires personnel to enter hazardous areas, potentially causing injury to operators. Furthermore, sample analysis is time-consuming, failing to meet the demands of online monitoring. Although gas sensors are deployed in industrial pipelines, their coverage area is typically limited, making them unsuitable for scenarios involving large-scale gas diffusion. Gas telemetry technology, on the other hand, can provide online early warnings of gas composition, regional distribution, and diffusion patterns from a distance, offering significant advantages in practical applications.

[0004] Gas telemetry imaging, based on different acquisition principles, is divided into hyperspectral gas imaging, multispectral gas imaging, and gas thermal imaging detection technology. Hyperspectral imaging instruments primarily utilize the FTIR (Fourier Transform Infrared) principle to acquire the infrared fingerprint spectrum of gases for remote gas identification and early warning. FTIR offers high sensitivity, can measure up to thousands of gas types, and has a long detection distance, but requires high-performance cooled detectors, typically costing over a million RMB. Multispectral gas imaging instruments employ filters or static spectroscopic structures, resulting in lower spectral resolution and typically measuring only a few dozen gas types. They use uncooled focal plane detectors for gas imaging, reducing costs while meeting industrial application requirements. Gas thermal imaging detection acquires plume images by detecting gas temperature. Its disadvantage is that it can only detect a single gas; to detect another gas, the pre-filter needs to be replaced, and gas warnings require manual observation. For example, the existing patent document CN105044758A, "Spectral Imaging Detector," includes a one-dimensional multi-element photodetector with a photodiode array. The photodiode array has a first upper row of photodiode pixels and a second lower row of photodiode pixels. The photodiode array is part of the photodetector. The scintillator array includes a first upper row scintillator pixel and a second lower row scintillator pixel. The first upper row scintillator pixel and the second lower row scintillator pixel are optically coupled to the first upper row photodiode pixel and the second lower row photodiode pixel, respectively. The photodetector also includes readout electronics, which are also part of the photodetector. Electrical traces interconnect the photodiode pixels with the readout electronics. As can be seen from the specification of this prior art, in this conventional spectral imaging detector, the radiation source, such as an X-ray tube, is supported by a rotating frame and rotates with it, emitting radiation. The source collimator calibrates the emitted radiation to form a radiation beam that is generally conical, fan-shaped, wedge-shaped, or other shaped, traversing the inspection area. It is known that the spectral detection data in the prior art is susceptible to internal radiation interference, resulting in low detection accuracy.

[0005] like Figure 1 and Figure 2 As shown, Telops, a Canadian company, developed the HyperCam series of infrared imaging spectrometers. With a large infrared focal plane and imaging spectral structure, it can simultaneously acquire spectral information from tens of thousands of spatial locations. Through high-speed parallel analysis and processing algorithms, it can directly obtain the spatial distribution of target clouds on a two-dimensional plane, greatly improving the capabilities of infrared remote sensing technology.

[0006] like Figure 3 and Figure 4As shown, addressing the technical bottlenecks of insufficient imaging capabilities of infrared band unit detectors and the high cost and limited bandwidth of area array detectors, the German company Bruker proposed a scanning imaging approach. By combining a high-speed Fourier transform interferometer module and a two-dimensional scanning mechanism, the spatial distribution of unit detection results can be reconstructed while ensuring bandwidth, enabling the imaging display of infrared spectral analysis results. A representative product is the company's SIGIS2.

[0007] In recent years, multispectral imaging technology using uncooled detectors has developed rapidly, offering advantages such as imaging capability, early warning capability, and reasonable price.

[0008] like Figure 5 and Figure 6 As shown, the Second-Sight series from the French company Bertin consists of an infrared thermal imager covering the spectral characteristics of the gas to be detected and a filter module. The filter module contains both reference and measurement filters. The Second-Sight series uses a 384×272 pixel uncooled infrared focal plane detector, operating in the 8–14 μm long-wavelength range. It employs an infrared long-pass filter instead of the narrow-band filters used in traditional optical gas detection, providing high infrared transmission, high detection accuracy, and sensitive alarms. It retains the imaging capabilities of a camera, enabling visualization of leaked gas clouds.

[0009] like Figure 7 and Figure 8 As shown, the GCI gas imaging telemetry system developed by Rebellion in the United States uses static spectrophotometry to acquire multispectral information of gases. The instrument uses vision to identify and quantify gas leaks online, with a detection range of up to 1.6 kilometers. This intelligent monitoring system integrates an advanced artificial intelligence-driven software platform. When a gas leak, fire, or safety issue is detected, the platform automatically sends an alarm to plant operators and provides detailed analysis. GCI has developed proprietary spectral imaging technology and detection algorithms to accurately locate leak sources, and it is applied to leak detection and monitoring in the petrochemical industry.

[0010] like Figure 9 and Figure 10As shown, the Gas FindIR series of products developed by FLIR Systems, Inc. is based on the infrared characteristic absorption of gas molecules. It achieves visualized gas measurement by adding an infrared narrow-bandpass filter with transmittance matching the wavelength of bond vibration or rotational energy transitions in gas molecules before the infrared focal plane array, and by cooling both the infrared focal plane array and the narrow-bandpass filter together to reduce internal infrared radiation. Different gases have different infrared characteristic absorption peak spectra; therefore, FLIR has developed different models for different gases. For example, the GF304 model, with a working spectral range of 10.3–10.7 μm, can measure sulfur hexafluoride; the GF300 model, with a working spectral range of 3.2–3.4 μm, can measure methane; and the GF346 model, with a working spectral range of 4.52–4.76 μm, can measure carbon monoxide, etc.

[0011] The disadvantages of gas thermal imaging detection are that it can only detect one type of gas at a time, the image display results still need to be manually interpreted, it does not have an early warning function, and its application scenarios are limited.

[0012] The advantages and disadvantages of the above gas imaging techniques are compared as follows:

[0013] Table 1. Classification and Comparison of Different Technologies

[0014]

[0015] The existing patent document CN113340425A, entitled "An Infrared Cloud Imager System for Space-to-Ground Laser Communication," includes an optical sensing module, a communication control module, and an algorithm unit. The optical sensing module consists of a microwave radiometric camera equipped with a wide-angle lens. The communication control module utilizes a microcomputer and adopts a server-client model. The algorithm unit includes a radiometric calibration algorithm module and a cloud optical depth algorithm module. The technical solution disclosed in this existing patent application utilizes scene remote sensing images and real-time temperature information, and obtains a corrected stable camera response through a focal plane array temperature correction coefficient. The corrected camera response value is converted into a radiometric image through laboratory blackbody radiation calibration. This existing technology, based on the cloud infrared grayscale image acquired from the field of view above the system's location, uses radiometric calibration and optical depth algorithm analysis to obtain an infrared radiation image of the sky and an optical depth distribution image of the clouds. However, this method, using the radiometric calibration principle, only addresses image detection and processing and cannot eliminate radiation interference generated within the system equipment.

[0016] The above comparison of different technologies shows that multispectral gas imaging technology offers a good balance between detection performance and price, and can meet the online detection needs of hazardous gases in industrial scenarios. However, the relevant technologies are currently mainly monopolized by foreign countries, and there are no mature products in China. Breakthroughs in gas imaging telemetry technology and detection algorithms are urgently needed.

[0017] In summary, existing technologies suffer from poor infrared calibration, leading to low accuracy in gas imaging telemetry and necessitating the use of active light sources. Summary of the Invention

[0018] The technical problem to be solved by this invention is how to solve the problem that the poor infrared calibration effect of the existing technology leads to low accuracy of gas imaging telemetry and the need to use an active light source.

[0019] This invention solves the above-mentioned technical problems by employing the following technical solution: A gas early warning and radiation calibration system based on an uncooled infrared camera includes:

[0020] The lens is positioned at a preset external light incident position to collect incident light reflected from external gas for gas imaging;

[0021] No fewer than two filters are mounted on a filter wheel to switch atmospheric window bands by rotating the filter wheel. The atmospheric window bands include the range [8μm, 14μm], and the filters are bandpass filters.

[0022] An uncooled infrared camera is mounted on the outside of a filter in the direction of the reflected incident light to acquire infrared images;

[0023] The gas imaging module is used to acquire the radiance of a single pixel of a filter using infrared images. The difference between two single pixels is calculated to obtain the radiance difference. The ratio of the radiance difference is processed to obtain the multispectral features of the target. The long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the short-wavelength cutoff wavelength of the reference filter are processed by preset logic to obtain the target multispectral representation data. The multispectral representation data is processed using the least squares method to obtain the gas concentration range value of the target gas. The gas imaging module is connected to the filter and the uncooled infrared camera.

[0024] The non-uniformity correction module is used to measure at least two differential temperature blackbodies under a preset temperature environment to obtain voltage values, calculate the average value of all pixels of the differential temperature blackbodies, calculate the radiative exitance of the differential temperature blackbodies using the Stefan-Boltzmann law to obtain response standard values ​​and bias standard values, calibrate the bias of each pixel, and use the least squares method to obtain the bias term calibration value of each pixel to calibrate the response of each pixel. It also processes the temperature average value and temperature average value of the differential temperature blackbodies to shift the pixel response to the standard value as the non-uniformity correction result. The non-uniformity correction module is connected to an uncooled infrared camera.

[0025] The temperature correction module is used to collect the lens temperature and the core temperature, measure no less than two temperature blackbody voltage values ​​respectively, process the pixel voltage value, lens temperature, response and core temperature according to the nonlinear model of pixel voltage value and lens core temperature, and (9) use the least squares method to calculate the core related correction coefficient, lens related correction coefficient and preset constant, so as to obtain the temperature correction result. The temperature correction module is connected to the non-uniform correction module.

[0026] The shutter hood correction module is used to rotate the shutter hood to a preset shutter hood position between the lens and the filter to measure the shutter voltage value. The corrected voltage value is obtained by subtracting the shutter voltage value from the externally measured voltage value. The shutter hood correction module is connected to the temperature correction module.

[0027] The radiometric calibration module is used to convert the camera's measured voltage values ​​into temperature values, and then process them to obtain radiometric calibration coefficients. A preset number of blackbodies are placed at the light-collecting position of the lens. The measurement system stabilizes the blackbodies' voltage values, extracts the blackbodies' blackbody region pixels, sets the temperature of the remaining blackbodies, obtains and processes the average value of the voltage difference between the temperatures, and then obtains the radiometric calibration coefficients for radiometric calibration. The radiometric calibration module is connected to the shutter correction module.

[0028] This invention acquires infrared images at different wavelengths and identifies and classifies gases based on their multispectral characteristics at different wavelengths. The system's radiation calibration converts voltage values ​​into temperature values. Addressing the problem of internal radiation variations and interference with multispectral data, which can lead to undetected gas targets or false alarms, this invention employs non-uniform correction to calibrate the response and bias of each pixel to the same value, corrects for temperature variations, and utilizes a light-shielding shutter in conjunction with a calibration model to obtain radiation calibration parameters, thus achieving instrument radiation calibration.

[0029] This invention acquires temperature data in different wavelength bands by correcting camera non-uniformity, internal instrument radiation, and converting voltage to temperature. The system offers advantages such as low cost, imaging capability, long-distance telemetry, no need for an active light source, and simultaneous monitoring of multiple gases. Furthermore, this invention utilizes the 8-14μm infrared atmospheric window band, based on the principle of passive infrared telemetry, using natural objects as a background, and requires no additional light source.

[0030] In a more specific technical solution, the gas imaging module includes:

[0031] The radiance processing unit is used to obtain the radiance of a pixel of a certain filter through the following logic processing:

[0032] L=∫[L BB (T B )τ Cloud +L BB (TCloud )(1-τ Cloud )]dλ

[0033] =∫{[L BB (T B )-L BB (T Cloud )]τ Cloud +L BB (T Cloud )}dλ

[0034] In the formula, T B It is the temperature of the background object; T Cloud It is the temperature of the cloud cluster; L BB It is the equivalent blackbody radiance; τ Cloud It is the cloud transmittance;

[0035] The radiance difference unit is used to calculate the radiance difference between two adjacent pixels where the cloud temperature and density are the same but the background temperature is different. The radiance difference unit is connected to the radiance processing unit.

[0036] ΔL=∫[L BB (T B1 )-L BB (T B2 )]τ Cloud dλ;

[0037] A multispectral feature unit is used to define the target filter as n and the reference filter as m, and to use the ratio of their radiance difference as the target multispectral feature. The multispectral feature unit is connected to the radiance difference unit.

[0038]

[0039] The multispectral representation unit is used to process the long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the reference filter short-wavelength cutoff wavelength using the following logic to obtain the target multispectral representation data. The multispectral representation unit is connected to the multispectral feature unit:

[0040]

[0041] In the formula, λ C It is the long-wavelength cutoff wavelength of the system; λ act and λ ref It is the short-wavelength cutoff wavelength of the target and reference filters;

[0042] The range length processing unit measures gases with different concentration range lengths, obtains coefficients a and b through regression, and processes the target multispectral representation data using the following least-squares regression logic to obtain the concentration range length value. The range length processing unit is connected to the multispectral representation unit.

[0043] C=a+bη nm .

[0044] This invention uses multiple broadband filters to obtain multispectral information of the gas, and realizes target component identification and early warning based on the multispectral information.

[0045] In a more specific technical solution, the non-uniformity correction module includes:

[0046] The blackbody voltage measurement unit is used to measure the voltage U of at least two blackbodies with different temperatures under a preset temperature environment. 1,(i,j) and U 2,(i,j) Where (i,j) represents the i-th row and j-th column;

[0047] The response bias processing unit is used to calculate the average value of all pixels of the differential temperature blackbody. and The radiative exitance Φ1 and Φ2 of the differential temperature blackbody are calculated using the Stefan-Boltzmann law to obtain the response standard value A and the bias standard value B. The response bias processing unit is connected to the blackbody voltage measurement unit.

[0048] The blackbody temperature mean difference processing unit is used to calibrate the bias of each pixel, using the average temperature of the blackbody at the first difference temperature. As the first standard value, the voltage value U of pixel (i,j) is obtained by processing it. 1,(i,j) The average temperature of the blackbody with the first difference temperature The first difference ΔU 1,(i,j) The average temperature of the blackbody with the second difference temperature As a second standard value, the voltage value U of pixel (i,j) is calculated. 2,(i,j) The average temperature of the blackbody with the second difference temperature The second difference ΔU 2,(i,j) The blackbody temperature mean difference processing unit is connected to the response bias processing unit;

[0049] The bias calibration processing unit is used to process the first difference ΔU using the least squares method. 1,(i,j) The second difference ΔU 2,(i,j) The average temperature of the first differential temperature blackbody And the average temperature of the second differential temperature blackbody The bias calibration value for each pixel is obtained to calibrate the response of each pixel. The bias calibration processing unit is connected to the blackbody temperature mean difference processing unit.

[0050] The response translation unit is used to process the average temperature of the blackbody with the first difference temperature. And the average temperature of the second differential temperature blackbody The response a is obtained as a standard value, and the response of pixel (i,j) is shifted to the standard value as the result of non-uniformity correction. The response shifting unit is connected to the bias calibration processing unit.

[0051] The image displayed by the present invention, which is directly based on voltage value, will have different gray levels depending on the pixel. By correcting for non-uniformity, the response of the pixels is shifted to the standard value, thereby improving the radiation calibration effect for the internal bias of the instrument.

[0052] In a more specific technical solution, the response bias processing unit calculates the radiative exitance Φ1 and Φ2 of the differential temperature blackbody using the following logic to obtain the response standard value A and the bias standard value B:

[0053]

[0054] In a more specific technical solution, the bias calibration processing unit processes the first difference ΔU using the following logic. 1,(i,j) The second difference ΔU 2,(i,j) The average temperature of the first differential temperature blackbody And the average temperature of the second differential temperature blackbody The bias calibration value for each pixel is obtained accordingly. The bias calibration processing unit is connected to the blackbody temperature mean difference processing unit.

[0055]

[0056] This is used to calibrate the response of each pixel.

[0057] In a more specific technical solution, the response translation unit processes the average temperature of the first differential temperature blackbody using the following logic. And the average temperature of the second differential temperature blackbody The response 'a' is obtained as a standard value, and the response of pixel (i,j) is shifted to the standard value. The response shifting unit is connected to the offset calibration processing unit.

[0058]

[0059] In a more specific technical solution, the temperature correction module includes:

[0060] The temperature acquisition unit is used to set different ambient temperatures. After the system stabilizes, it acquires the lens temperature and the body temperature.

[0061] The voltage measurement unit is used to measure the voltage values ​​of at least two blackbody temperatures at the same temperature. The voltage measurement unit is connected to the temperature acquisition unit.

[0062] The temperature correction processing unit processes the voltage value U of pixel (i,j) based on the following pixel voltage value and a lens movement temperature nonlinear model. (i,j) Lens temperature T C Response and movement temperature T s :

[0063]

[0064] In the formula, the subscript s represents the mechanism, C represents the lens; a represents the mechanism-related correction coefficient; b represents the lens-related correction coefficient; c is the preset constant. (9) The mechanism-related correction coefficient, lens-related correction coefficient and preset constant are calculated using the least squares method, and temperature correction is performed accordingly. The temperature correction processing unit is connected to the temperature acquisition unit and the voltage measurement unit.

[0065] This invention reduces the interference of lens temperature and core temperature on the electrical signals received by the infrared camera by processing the internal temperature and electrical signal data of the instrument through temperature correction, thus avoiding the impact of temperature changes on telemetry and early warning.

[0066] In a more specific technical solution, the shutter correction module includes:

[0067] A light-blocking shutter is positioned on the straight line of the incident light source. It is used to move the shutter into the field of view of the uncooled infrared camera to measure the shutter voltage value U. sh When it is necessary to measure external radiation, the shutter does not obstruct the field of view;

[0068] The voltage data processing unit is used to subtract the shutter voltage value from the externally measured voltage value: U' = UU sh The corrected voltage value is obtained based on this data, and the voltage data processing unit is connected to the shutter.

[0069] This invention reduces the interference of lens temperature changes and filter reflection temperature on the measured voltage value by rotating the shutter in conjunction with voltage data processing, and uses the shutter to correct the internal radiation of the instrument.

[0070] In a more specific technical solution, the radiation calibration module includes:

[0071] The blackbody voltage measurement unit is used to place a preset number of blackbodies at the light-collecting position of the lens, set the temperature of the blackbody, and measure the stable blackbody voltage value of the system.

[0072] The voltage averaging processing unit is used to extract the blackbody region pixels of the blackbody and calculate the voltage average value U based on them. The voltage averaging processing unit is connected to the blackbody voltage value measurement unit.

[0073] The blackbody group processing unit sets the temperature of at least two other blackbodies and cyclically executes the steps corresponding to the blackbody voltage value measurement unit and the voltage average value processing unit to obtain the average voltage value of the temperature difference. The blackbody group processing unit is connected to the voltage average value processing unit.

[0074] The radiation calibration unit processes the average value of the temperature difference voltage according to the following correction model to obtain radiation calibration coefficients for radiation calibration:

[0075] T o =d3U 3 +d2U 2 +d1U+d0.

[0076] In more specific technical solutions, gas early warning and radiation calibration methods based on uncooled infrared cameras include:

[0077] S1. Collect incident light reflected from external gas to perform gas imaging;

[0078] S2. Rotate the filter wheel to switch the atmospheric window band, wherein the atmospheric window band includes the range [8μm, 14μm], and the filter is a bandpass filter;

[0079] S3. Acquire infrared images using an uncooled infrared camera;

[0080] S4. Obtain the radiance of a single pixel of a filter using an infrared image. Calculate the difference between two single pixels to obtain the radiance difference. Calculate the ratio of the radiance difference to obtain the target multispectral features. Process the long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the reference filter short-wavelength cutoff wavelength using preset logic to obtain the target multispectral representation data. Process the multispectral representation data using the least squares method to obtain the gas concentration range value of the target gas.

[0081] S5. Under a preset temperature environment, measure no less than two differential temperature blackbodies to obtain voltage values. Calculate the average value of all pixels of the differential temperature blackbodies. Calculate the radiative exitance of the differential temperature blackbodies using the Stefan-Boltzmann law to obtain the standard response value and the standard bias value. Calibrate the bias of each pixel. Use the least squares method to obtain the bias calibration value of each pixel to calibrate the response of each pixel. Process the temperature average value and temperature average value of the differential temperature blackbodies to shift the pixel response to the standard value as the non-uniformity correction result.

[0082] S6. Collect and obtain the lens temperature and the core temperature, and measure no less than two temperature blackbody voltage values ​​respectively. Process the pixel voltage value, lens temperature, response and core temperature according to the nonlinear model of pixel voltage value and lens core temperature, and (9) use the least squares method to calculate the core related correction coefficient, lens related correction coefficient and preset constant, so as to obtain the temperature correction result.

[0083] S7. Rotate the shutter to the preset shutter position between the lens and the filter to measure the shutter voltage value. Subtract the shutter voltage value from the externally measured voltage value to obtain the corrected voltage value.

[0084] S8. Convert the camera's measured voltage value into a temperature value, process it to obtain the radiation calibration coefficient, place a preset number of blackbodies at the light acquisition position of the lens, measure the stable blackbodies voltage value, extract the blackbodies' blackbody region pixels, set the temperature of the remaining blackbodies, obtain and process the average value of the difference in temperature voltage, and obtain the radiation calibration coefficient for radiation calibration.

[0085] Compared with existing technologies, this invention has the following advantages: This invention acquires infrared images at different wavelengths and identifies and classifies gases based on their multispectral characteristics at different wavelengths. The system's radiation calibration converts voltage values ​​into temperature values. Addressing the problem of internal radiation variations and interference with multispectral data, which can lead to undetected gas targets or false alarms, this invention employs non-uniform correction to calibrate the response and bias of each pixel to the same value, corrects for temperature variations, and utilizes a light-shielding shutter in conjunction with a calibration model to obtain radiation calibration parameters, thus achieving instrument radiation calibration.

[0086] This invention acquires temperature data in different wavelength bands by correcting camera non-uniformity, internal instrument radiation, and converting voltage to temperature. The system offers advantages such as low cost, imaging capability, long-distance telemetry, no need for an active light source, and simultaneous monitoring of multiple gases. Furthermore, this invention utilizes the 8-14μm infrared atmospheric window band, based on the principle of passive infrared telemetry, using natural objects as a background, and requires no additional light source.

[0087] This invention uses multiple broadband filters to obtain multispectral information of the gas, and realizes target component identification and early warning based on the multispectral information.

[0088] The image displayed by the present invention, which is directly based on voltage value, will have different gray levels depending on the pixel. By correcting for non-uniformity, the response of the pixels is shifted to the standard value, thereby improving the radiation calibration effect for the internal bias of the instrument.

[0089] This invention reduces the interference of lens temperature and core temperature on the electrical signals received by the infrared camera by processing the internal temperature and electrical signal data of the instrument through temperature correction, thus avoiding the impact of temperature changes on telemetry and early warning.

[0090] This invention reduces the interference of lens temperature changes and filter reflections on the measured voltage value by rotating the shutter in conjunction with voltage data processing. It also utilizes the shutter to correct for internal radiation within the instrument. This invention solves the technical problems of poor infrared calibration in existing technologies, which leads to low accuracy in gas imaging telemetry and necessitates the use of an active light source. Attached Figure Description

[0091] Figure 1 This is a picture of the HyperCam imaging spectrometer.

[0092] Figure 2 This is a schematic diagram of methane detection results in a swamp area using the HyperCam imaging spectrometer.

[0093] Figure 3 This is a schematic diagram of the SIGIS2 prototype in operation.

[0094] Figure 4 This is a schematic diagram of the SIGIS2 scanning imaging results;

[0095] Figure 5 Image of the Second-Sight instrument;

[0096] Figure 6 This is a schematic diagram of the imaging detection results from the Second-Sight instrument;

[0097] Figure 7 Image of the actual GCI instrument;

[0098] Figure 8 This is a schematic diagram of the imaging detection results;

[0099] Figure 9 This is a picture of the GF300 instrument.

[0100] Figure 10 For imaging detection results;

[0101] Figure 11 This is a schematic diagram of the target filter and reference filter in Embodiment 1 of the present invention;

[0102] Figure 12 This is a schematic diagram of the gas early warning and radiation calibration system based on an uncooled infrared camera according to Embodiment 1 of the present invention;

[0103] Figure 13 This is a schematic diagram of the basic radiation calibration process in Embodiment 1 of the present invention;

[0104] Figure 14 This is a schematic diagram of the specific process for non-uniformity correction in Embodiment 1 of the present invention;

[0105] Figure 15 This is a schematic diagram of the specific temperature calibration process in Embodiment 1 of the present invention;

[0106] Figure 16 This is a schematic diagram of the specific process of shutter correction in Embodiment 1 of the present invention;

[0107] Figure 17 This is a schematic diagram of the specific process of radiation calibration in Embodiment 1 of the present invention. Detailed Implementation

[0108] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0109] Example 1

[0110] The gas early warning and radiation calibration system based on an uncooled infrared camera provided by this invention adopts the following gas detection principle:

[0111] like Figure 11 As shown, the infrared band is divided into different bands using a bandpass filter 3 according to the different characteristic bands of the gas. Based on the different characteristic bands, there are target filters and reference filters. The target filter covers the target characteristic band, while the reference filter has no target characteristics.

[0112] The principle of target recognition is as follows:

[0113] The radiance of a single pixel in a given filter is:

[0114] L=∫[L BB (T B )τ Cloud +L BB (T Cloud )(1-τ Cloud )]dλ

[0115] =∫{[L BB (T B )-L BB (T Cloud )]τ Cloud +L BB (T Cloud )}dλ (1)

[0116] In the formula T B It is the temperature of the background object; T Cloud It is the temperature of the cloud cluster; L BB It is the equivalent blackbody radiance; τ Cloud It is the cloud transmittance.

[0117] Assuming the cloud temperature and density are the same for these two different pixels, but the background temperature is different, then the difference between two adjacent pixels is used to obtain the radiance difference:

[0118] ΔL=∫[L BB (T B1 )-L BB (T B2 )]τ Cloud dλ (2)

[0119] Assuming the target filter is numbered n and the reference filter is numbered m, the ratio of the two is the target multispectral characteristic:

[0120]

[0121] Substituting equation (2) into equation (3), we get:

[0122]

[0123] In the formula λ C It is the long-wavelength cutoff wavelength of the system; λ act and λ ref It is the short-wavelength cutoff wavelength of the target and reference filters.

[0124] The concentration range of the target gas can be obtained through least squares regression:

[0125] C=a+bη nm (5)

[0126] By measuring gases with different concentration ranges in the laboratory and obtaining coefficients a and b through regression, the gas concentration range can be calculated.

[0127] like Figure 12 As shown, the gas early warning and radiation calibration system based on an uncooled infrared camera 5 provided by the present invention includes:

[0128] The system structure is shown in the figure below. Lens 1 collects external light and enables clear imaging of objects; shutter 2 measures the internal radiation of the instrument. When measuring internal radiation, shutter 2 rotates between lens 1 and bandpass filter 3. When measuring external radiation, shutter 2 does not obstruct the field of view; bandpass filter wheel 4 carries several bandpass filters 3, which can be switched between different bands by rotation; uncooled infrared camera 5 acquires infrared images.

[0129] The system acquires infrared images at different wavelengths and identifies and classifies gases based on their multispectral characteristics at different wavelengths. The system's radiation calibration involves converting voltage values ​​into temperature values. Due to variations in internal instrument radiation and interference with multispectral data, gas targets may fail to be detected or false alarms may occur. Radiation calibration is a critical step in the measurement process.

[0130] like Figure 13 As shown, in the calibration process of this invention, the number of pixels in the uncooled infrared camera 5 is set to M×N, and the number of bandpass filters 3 is K. For the k-th bandpass filter 3, the system performs radiation calibration according to the following procedure:

[0131] S1, Non-uniform linear correction;

[0132] S2, Temperature Correction;

[0133] S3, shutter speed correction;

[0134] S4, Radiation Calibration.

[0135] (1) Non-uniformity correction

[0136] In an uncooled infrared camera 5, each pixel has a different response and bias, resulting in images displayed directly as voltage values ​​exhibiting different grayscale levels due to pixel differences. Non-uniformity correction calibrates the response and bias of each pixel to the same value.

[0137] Measure the temperature voltage values ​​of two blackbody bodies to obtain standard values ​​for the response and bias.

[0138] like Figure 14 As shown, in this embodiment, the non-uniformity correction step S1 further includes the following steps:

[0139] S11. Under normal indoor temperature conditions, measure the voltage U of two blackbodies at different temperatures. 1,(i,j) and U 2,(i,j) , where (i,j) represents the i-th row and j-th column.

[0140] S12. Calculate the average value of all pixels. and Calculate the blackbody radiative exitance Φ1 and Φ2 at the two temperatures using the Stefan-Boltzmann law, and substitute them into the formula.

[0141]

[0142] A and B are calculated as standard values ​​for the response and bias.

[0143] S13. Calibrate the offset of each pixel.

[0144] The process is as follows:

[0145] The average value of blackbody temperature 1 As a standard value, the voltage value U of pixel (i,j) is calculated. 1,(i,j) and The difference ΔU 1,(i,j) ; using the average value of blackbody temperature 2 As a standard value, the voltage value U of pixel (i,j) is calculated. 2,(i,j) and The difference ΔU 2,(i,j) Using the least squares method, calculate the calibration value for the bias term of each pixel:

[0146]

[0147] The response of each pixel is calibrated.

[0148] Using the average response 'a' as the standard value, the response of pixel (i,j) is shifted to the standard value, i.e.:

[0149]

[0150] (2) Temperature correction

[0151] The electrical signals received by the uncooled infrared camera 5 are affected by the combined temperature of the lens and the camera body, causing changes in the sensitivity and dark current of the electrical signals, which require correction for temperature changes.

[0152] Voltage value U of pixel (i,j) (i,j) With lens temperature T C and movement temperature T s The nonlinear model is as follows:

[0153]

[0154] In the formula, the subscript s represents the camera movement, C represents lens 1; a represents the correction coefficient related to the camera movement; b represents the correction coefficient related to lens 1; and c is a constant.

[0155] like Figure 15 As shown, step S2 of temperature correction also includes the following steps:

[0156] S21. Set different ambient temperatures respectively. After the system stabilizes, obtain the lens temperature and the core temperature.

[0157] S22. Measure the blackbody voltage values ​​at three different temperatures at the same temperature.

[0158] S23. Based on the nonlinear model (9), calculate the coefficients a, b, and c using the least squares method.

[0159] (3) Light-blocking shutter correction

[0160] Changes in internal radiation of the instrument (such as temperature changes in lens 1 and temperature of the movement reflected by bandpass filter 3) cause changes in the measured voltage value. The internal radiation of the instrument is corrected by using shutter 2.

[0161] like Figure 16 As shown, in this embodiment, step S3 of the shutter correction for light blocking further includes the following steps:

[0162] S31. Turn the shutter to be within the camera's field of view and measure the shutter voltage value U. sh ;

[0163] S32. Subtract the shutter speed value from the external measurement value, U' = UU sh The corrected voltage value is obtained.

[0164] (4) Radiation calibration

[0165] After the above correction process, the radiometric calibration converts the voltage values ​​measured by the uncooled infrared camera 5 into temperature values. The radiometric calibration model is as follows:

[0166] T o =d3U 3 +d2U 2 +d1U+d0 (10)

[0167] like Figure 17 As shown, the calibration step S4 of the radiation calibration also includes the following steps:

[0168] S41. Place the blackbody at the front of the instrument, set the blackbody temperature, and measure the blackbody voltage value after the system stabilizes.

[0169] S42. Select pixels in the blackbody region and calculate the average voltage U;

[0170] S43. Set two other different blackbody temperatures and repeat steps S41 and S42 above.

[0171] S44. The coefficient d is calculated based on the radiation calibration correction model.

[0172] In summary, this invention acquires infrared images at different wavelengths and identifies and classifies gases based on their multispectral characteristics at different wavelengths. The system's radiation calibration converts voltage values ​​into temperature values. Addressing the problem of internal radiation variations and interference with multispectral data, which can lead to undetected gas targets or false alarms, this invention employs non-uniform correction to calibrate the response and bias of each pixel to the same value, corrects for temperature variations, and utilizes a light-shielding shutter in conjunction with a calibration model to obtain radiation calibration parameters, thus achieving instrument radiation calibration.

[0173] This invention acquires temperature data in different wavelength bands by correcting camera non-uniformity, internal instrument radiation, and converting voltage to temperature. The system offers advantages such as low cost, imaging capability, long-distance telemetry, no need for an active light source, and simultaneous monitoring of multiple gases. Furthermore, this invention utilizes the 8-14μm infrared atmospheric window band, based on the principle of passive infrared telemetry, using natural objects as a background, and requires no additional light source.

[0174] This invention uses multiple broadband filters to obtain multispectral information of the gas, and realizes target component identification and early warning based on the multispectral information.

[0175] The image displayed by the present invention, which is directly based on voltage value, will have different gray levels depending on the pixel. By correcting for non-uniformity, the response of the pixels is shifted to the standard value, thereby improving the radiation calibration effect for the internal bias of the instrument.

[0176] This invention reduces the interference of lens temperature and core temperature on the electrical signals received by the infrared camera by processing the internal temperature and electrical signal data of the instrument through temperature correction, thus avoiding the impact of temperature changes on telemetry and early warning.

[0177] This invention reduces the interference of lens temperature changes and filter reflections on the measured voltage value by rotating the shutter in conjunction with voltage data processing. It also utilizes the shutter to correct for internal radiation within the instrument. This invention solves the technical problems of poor infrared calibration in existing technologies, which leads to low accuracy in gas imaging telemetry and necessitates the use of an active light source.

[0178] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A gas early warning and radiation calibration system based on an uncooled infrared camera, characterized in that, The system includes: The lens is positioned at a preset external light incident position to collect incident light reflected from external gas for gas imaging; Not less than two filters are mounted on a filter wheel to switch atmospheric window bands by rotating the filter wheel, wherein the atmospheric window band includes the range [8μm, 14μm], and the filter is a bandpass filter; An uncooled infrared camera is mounted on the filter in the direction of the incident light reflected by the external gas to acquire infrared images; A gas imaging module is used to acquire the radiance of a single pixel of the filter using the infrared image, calculate the difference between two single pixels to obtain the radiance difference, and perform ratio processing based on the radiance difference to obtain the target multispectral features; it processes the long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the short-wavelength cutoff wavelength of the reference filter using preset logic to obtain target multispectral representation data, and processes the multispectral representation data using the least squares method to obtain the gas concentration range value of the target gas. The gas imaging module is connected to the filter and the uncooled infrared camera. The gas imaging module includes: The radiance processing unit is used to obtain the radiance of a pixel of a certain filter through the following logic processing: In the formula, It is the temperature of the background object; It's the temperature of the cloud cluster; It is the equivalent blackbody radiance; It is the cloud transmittance; A radiance difference unit is used to calculate the radiance difference between two adjacent pixels where the cloud temperature and density are the same, but the background object temperature is different. The radiance difference unit is connected to the radiance processing unit. ； A multispectral feature unit is used to define the target filter as n and the reference filter as m, and to use the ratio of their radiance difference as the target multispectral feature. The multispectral feature unit is connected to the radiance difference unit. A multispectral representation unit is used to process the long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the reference filter short-wavelength cutoff wavelength using the following logic to obtain target multispectral representation data. The multispectral representation unit is connected to the multispectral feature unit. In the formula, It is the long-wavelength cutoff wavelength of the system; and It is the short-wavelength cutoff wavelength of the target filter and the reference filter; A range length processing unit is used to measure gases with different concentration range lengths, obtain coefficients a and b through regression, and then process the target multispectral representation data according to the following least squares regression logic to obtain the concentration range length value. The range length processing unit is connected to the multispectral representation unit. A non-uniformity correction module is used to measure at least two differential temperature blackbodies under a preset temperature environment to obtain voltage values, calculate the average value of all pixels of the differential temperature blackbodies, calculate the radiative exitance of the differential temperature blackbodies using the Stefan-Boltzmann law to obtain response standard values ​​and bias standard values, calibrate the bias of each pixel, and obtain the bias standard value of each pixel using the least squares method to calibrate the response of each pixel, process the average temperature of the first differential temperature blackbodies and the average temperature of the second differential temperature blackbodies, and shift the response of the pixel to the response standard value as the non-uniformity correction result. The non-uniformity correction module is connected to the uncooled infrared camera. The temperature correction module is used to acquire lens temperature and core temperature, and measure no less than two temperature blackbody voltage values ​​respectively. Based on the pixel voltage value and lens core temperature nonlinear model, the voltage value of the pixel, the lens temperature, the response and the core temperature are processed. The nonlinear model uses the least squares method to calculate the core-related correction coefficient, the lens-related correction coefficient and the preset constant, so as to obtain the temperature correction result. The temperature correction module is connected to the non-uniformity correction module. The temperature calibration module includes: The temperature acquisition unit is used to set different ambient temperatures. After the system stabilizes, it acquires the lens temperature and the body temperature. A voltage measurement unit is used to measure at least two temperature blackbody voltage values ​​at the same temperature, and the voltage measurement unit is connected to the temperature acquisition unit; The temperature correction processing unit processes the pixel based on the following pixel voltage value and a nonlinear model of lens movement temperature. voltage value The lens temperature The response and the movement temperature : In the formula, the subscript s represents the movement, C represents the lens; a represents the movement-related correction coefficient; b represents the lens-related correction coefficient; c is a preset constant. The nonlinear model uses the least squares method to calculate the movement-related correction coefficient, the lens-related correction coefficient, and the preset constant, and performs temperature correction accordingly. The temperature correction processing unit is connected to the temperature acquisition unit and the voltage measurement unit. The shutter hood correction module is used to rotate the shutter hood to a preset shutter hood position between the lens and the filter to measure the shutter voltage value. The corrected voltage value is obtained by subtracting the shutter voltage value from the externally measured voltage value. The shutter hood correction module is connected to the temperature correction module. The radiation calibration module is used to convert the camera's measured voltage value into a temperature value, and then process it to obtain the radiation calibration coefficient. A preset number of black bodies are placed at the light acquisition position of the lens. The measurement system stabilizes the black body voltage value, extracts the black body region pixels of the black body, sets the temperature of the remaining black bodies, obtains and processes the average value of the difference temperature voltage, and then obtains the radiation calibration coefficient for radiation calibration. The radiation calibration module is connected to the light-blocking shutter correction module. The radiation calibration module includes: A blackbody voltage measurement unit is used to place a preset number of blackbodies at the light acquisition position of the lens, set the temperature of the blackbody, and measure the stable blackbody voltage value of the system accordingly. The voltage averaging unit is used to extract the pixels of the blackbody region of the blackbody and calculate the voltage average value accordingly. The voltage average value processing unit is connected to the blackbody voltage value measurement unit; The blackbody group processing unit sets the temperature of at least two other blackbodies to cyclically execute the steps corresponding to the blackbody voltage value measurement unit and the voltage average value processing unit, thereby obtaining the average voltage value of the temperature difference. The blackbody group processing unit is connected to the voltage average value processing unit. The radiation calibration unit processes the average value of the differential temperature voltage according to the following correction model to obtain the radiation calibration coefficients for radiation calibration: 。 2. The gas early warning and radiation calibration system based on an uncooled infrared camera according to claim 1, characterized in that, The non-uniformity correction module includes: The blackbody voltage measurement unit is used to measure the voltage of at least two blackbodies with different temperatures under a preset temperature environment. and ,in, Indicates the first Line 1 Column of pixels; A response bias processing unit is used to calculate the average value of all pixels of the differential temperature blackbody. and The radiative exitance of the differential temperature blackbody was calculated using the Stefan-Boltzmann law. and To obtain the standard response value and bias standard value The response bias processing unit is connected to the blackbody voltage measurement unit; The blackbody temperature mean difference processing unit is used to calibrate the bias of each pixel, using the average temperature of the blackbody at the first difference temperature. The pixel is obtained by processing it based on the first standard value. voltage value The average temperature of the blackbody with the first temperature difference First difference The average temperature of the blackbody with the second difference temperature As a second standard value, the pixel is calculated. voltage value The average temperature of the second differential temperature blackbody The second difference The blackbody temperature mean difference processing unit is connected to the response bias processing unit; The bias calibration processing unit is used to process the first difference using the least squares method. The second difference The average temperature of the first differential temperature blackbody and the average temperature of the second differential temperature blackbody The bias standard value of each pixel is obtained accordingly to calibrate the response of each pixel. The bias calibration processing unit is connected to the blackbody temperature mean difference processing unit. The response translation unit is used to process the average temperature of the first differential temperature blackbody. and the average temperature of the second differential temperature blackbody Based on this, a response was obtained. As a response standard value, to the pixel The response is shifted to the standard response value as a result of non-uniformity correction, and the response shifting unit is connected to the bias calibration processing unit.

3. The gas early warning and radiation calibration system based on an uncooled infrared camera according to claim 2, characterized in that, The response bias processing unit calculates the radiative exitance of the differential temperature blackbody using the following logic. and To obtain the standard response value and bias standard value : 。 4. The gas early warning and radiation calibration system based on an uncooled infrared camera according to claim 2, characterized in that, The bias calibration processing unit processes the first difference using the following logic. The second difference The average temperature of the first differential temperature blackbody and the average temperature of the second differential temperature blackbody The bias standard value for each pixel is obtained accordingly, and the bias calibration processing unit is connected to the blackbody temperature mean difference processing unit. To calibrate the response of each pixel.

5. The gas early warning and radiation calibration system based on an uncooled infrared camera according to claim 2, characterized in that, The response translation unit processes the average temperature of the first differential temperature blackbody using the following logic. and the average temperature of the second differential temperature blackbody Based on this, a response was obtained. As a response standard value, to the pixel The response is shifted to the standard response value, and the response shifting unit is connected to the bias calibration processing unit: 。 6. The gas early warning and radiation calibration system based on an uncooled infrared camera according to claim 1, characterized in that, The light-blocking shutter correction module includes: A light-blocking shutter, positioned along the line of incident light, is used to rotate the shutter into the field of view of the uncooled infrared camera to measure the shutter voltage value. When it is necessary to measure external radiation, the shutter does not obstruct the field of view; A voltage data processing unit is used to subtract the shutter voltage value from an externally measured voltage value. The corrected voltage value is obtained based on this data, and the voltage data processing unit is connected to the light-blocking shutter.

7. A gas early warning and radiation calibration method based on an uncooled infrared camera, based on the gas early warning and radiation calibration system based on an uncooled infrared camera according to any one of claims 1 to 6, characterized in that, The method includes: S1. Collect incident light reflected from external gas to perform gas imaging; S2. Rotate the filter wheel to switch the atmospheric window band, wherein the atmospheric window band includes the range [8μm, 14μm], and the filter is a bandpass filter; S3. Acquire infrared images using an uncooled infrared camera; S4. Obtain the radiance of a single pixel of the filter using the infrared image, calculate the difference between two single pixels to obtain the radiance difference, and perform ratio processing based on the radiance difference to obtain the target multispectral features; process the long-wavelength cutoff wavelength, the target short-wavelength cutoff wavelength, and the short-wavelength cutoff wavelength of the reference filter using preset logic to obtain the target multispectral representation data; process the multispectral representation data using the least squares method to obtain the gas concentration range value of the target gas; S5. Under a preset temperature environment, measure no less than two differential temperature blackbodies to obtain voltage values, calculate the average value of all pixels of the differential temperature blackbodies, calculate the radiative exitance of the differential temperature blackbodies using the Stefan-Boltzmann law to obtain response standard values ​​and bias standard values, calibrate the bias of each pixel, and obtain the bias standard value of each pixel using the least squares method to calibrate the response of each pixel, process the average temperature of the first differential temperature blackbodies and the average temperature of the second differential temperature blackbodies, and shift the response of the pixel to the response standard value as the non-uniformity correction result. S6. Acquire lens temperature and core temperature, and measure no less than two temperature blackbody voltage values ​​respectively. Process the pixel voltage value, lens temperature, response and core temperature according to the nonlinear model of pixel voltage value and lens core temperature. The nonlinear model uses the least squares method to calculate the core related correction coefficient, lens related correction coefficient and preset constant, so as to obtain the temperature correction result. S7. Rotate the shutter to the preset shutter position between the lens and the filter to measure the shutter voltage value. Subtract the shutter voltage value from the externally measured voltage value to obtain the corrected voltage value. S8. Convert the camera's measured voltage value into a temperature value, process it to obtain a radiation calibration coefficient, place a preset number of blackbodies at the light acquisition position of the lens, measure the stable blackbodies voltage value, extract the blackbodies' blackbodies region pixels, set the temperature of the remaining blackbodies, obtain and process the average value of the difference temperature voltage, and obtain the radiation calibration coefficient for radiation calibration.

Images