On-orbit real-time surface temperature inversion method based on long-wave infrared image

Abstract

The invention provides an on-orbit real-time surface temperature inversion method based on a long-wave infrared image, and the method comprises the steps: obtaining the spectral radiance of a two-spectrum long-wave infrared image according to a DN value and a calibration coefficient of the two-spectrum long-wave infrared image obtained in an on-orbit manner; according to the spectral radiance of the two-spectral-segment long-wave infrared image, determining the brightness temperature of the two-spectral-segment long-wave infrared image; according to the DN value of the visible light image, the DN value of the near-infrared image and the calibration coefficient, determining the reflectivity of the visible light red spectrum section image and the near-infrared image; the NDVI index is determined according to the reflectivity of the visible light red spectrum section image and the near-infrared image; according to the reflectivity, the NDVI index and the vegetation coverage of the visible light red spectrum section image, the emissivity of the two-spectrum-section long-wave infrared image is obtained; and determining the surface temperature according to the emissivity and brightness temperature of the two-spectrum long-wave infrared image. According to the method, the surface temperature is inversed in real time based on various image data obtained in orbit, and the requirement of satellite users for the timeliness of satellite images is met.

Description

Technical Field

[0001] The present invention belongs to the technical field of surface temperature monitoring, and particularly relates to a method for on-orbit real-time inversion of surface temperature based on long-wave infrared images. Background Art

[0002] In the field of infrared quantitative remote sensing, infrared temperature products are one of the important products. Infrared temperature products can play important functions in many fields, such as research on the Earth's radiation budget, research on climate change trends, research on the urban heat island effect, etc. In addition, infrared temperature inversion can be used to monitor high-temperature ground objects such as high-temperature smoke points, thermal power, and high-temperature emissions from factories, and its monitoring results directly contribute to environmental protection supervision and disaster prevention and mitigation operations. However, the above-mentioned operations often require high real-time performance, and usually require satellites to have on-orbit imaging and anomaly warning capabilities. Therefore, it is required that satellites have the ability to generate temperature products in real time on orbit.

[0003] Limited by the complexity of the space environment, the FPGA+DSP scheme is often used as the core computing hardware of satellites to ensure the on-orbit stability and long-life operation of hardware resources. In recent years, due to the further development of semiconductor technology, new hardware architectures such as ARM+GPU have emerged. Affected by the progress of computing resources, satellite on-orbit processing has become a research hotspot.

[0004] The progress of on-board processing capabilities makes it possible to obtain surface temperature in a timely manner on the ground. The traditional method uses the DN value of infrared (mid-wave, long-wave) images for anomaly interpretation. This method is widely used in anomaly detection of high-temperature ground targets such as fire points and smoke points, but it is relatively ineffective in scenarios facing normal-temperature targets and requiring absolute radiation accuracy. The task of anomaly extraction of normal-temperature targets, such as sewage discharge monitoring, often carries out anomaly identification based on physical quantities such as temperature. Therefore, it is necessary to develop the ability to obtain surface temperature on orbit to meet the needs of anomaly extraction of surface temperature targets. In addition, obtaining surface temperature on orbit can still take into account the anomaly identification task of high-temperature targets.

[0005] How to apply on-orbit parallel computing resources and how to adaptively transform traditional ground algorithms directly affect the computing efficiency of software algorithms and the on-orbit processing capabilities of satellites. Therefore, the design of the on-orbit temperature inversion algorithm architecture is one of the key issues that current technical personnel are concerned about. Summary of the Invention

[0006] In order to overcome the deficiencies in the prior art, the inventor of the present invention has conducted intensive research and provided a method for on-orbit real-time inversion of surface temperature based on long-wave infrared images, which uses various spectral images obtained on orbit to obtain surface temperature data with high precision in real time; uses a parallel architecture hardware that can work stably on orbit, and through targeted simplification of mathematical formulas in the calculation process, division of calculation blocks, and reorganization of processes, realizes rapid on-orbit inversion of surface temperature while ensuring calculation accuracy.

[0007] The technical solution provided by the present invention is as follows:

[0008] In a first aspect, a method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images includes:

[0009] Obtain the spectral radiance of two spectral bands of long-wave infrared images based on the DN values and calibration coefficients of the two spectral bands of long-wave infrared images acquired on orbit;

[0010] Determine the brightness temperature of the two spectral bands of long-wave infrared images based on the spectral radiance of the two spectral bands of long-wave infrared images;

[0011] Determine the reflectivities of the visible light red spectral band image and the near-infrared image based on the DN value of the visible light image, the DN value of the near-infrared image, and the calibration coefficient;

[0012] Determine the NDVI index through the reflectivities of the visible light red spectral band image and the near-infrared image;

[0013] Based on the reflectivity of the visible light red spectral band image, the NDVI index, and the vegetation coverage P v , obtain the emissivity of the two spectral bands of long-wave infrared images;

[0014] Determine the land surface temperature based on the emissivity and brightness temperature of the two spectral bands of long-wave infrared images.

[0015] In a second aspect, a device for on-orbit real-time inversion of land surface temperature based on long-wave infrared images includes:

[0016] One or more processors;

[0017] A storage device for storing one or more programs,

[0018] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images described in the first aspect.

[0019] In a third aspect, a readable storage medium stores a computer program, and when the program is executed by a processor, the method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images described in the first aspect is implemented.

[0020] In a fourth aspect, a computer program product includes: a computer program (which can also be referred to as code or instructions), and when the computer program is run, it executes the method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images described in the first aspect.

[0021] A method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images provided by the present invention has the following beneficial effects:

[0022] (1) A method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images provided by the present invention is based on on-orbit acquired long-wave infrared images, visible red spectral band images, and near-infrared image data, and can real-time invert the land surface temperature, meeting the requirements of satellite users for the timeliness of satellite images;

[0023] (2) When there are multiple processing units, a method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images provided by the present invention improves the processing efficiency and reduces the occupancy rate of the parallel computing hardware cache during each calculation step by establishing a parallel processing unit hardware architecture to simultaneously process two-spectrum long-wave infrared data and visible + near-infrared data, implementing process reorganization, or simultaneously processing the operations in each step and splitting the calculation process;

[0024] (3) A method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images provided by the present invention changes the Planck calculation process from a logarithmic process to a radical process, uses the bisection method to solve the square root value, and can reduce the calculation time compared with calling functions to calculate logarithms and radicals under the condition of considering thread divergence, improving the calculation efficiency during on-orbit application;

[0025] (4) A method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images provided by the present invention is implemented using the C model of the OpenVX API and can be implemented on the OBT Yulong810A chip. Compared with the current embedded modules Combricon 1H8, NVIDIA Jetson TX2, and NVIDIA AGX, the OBT module has higher reliability and the ability to work stably for a long time in the space environment. Description of the Drawings

[0026] Figure 1 is the flow chart of the method for on-orbit real-time inversion of land surface temperature of the present invention;

[0027] Figure 2 is the spectral radiance map given by using the calibration parameters for the long-wave infrared data in the example simulation of the present invention;

[0028] Figure 3 is the brightness temperature data map generated from the long-wave infrared data in the example simulation of the present invention;

[0029] Figure 4 is the emissivity map generated using the NDVI method in the example simulation of the present invention;

[0030] Figure 5 is the calculation result map of the two parts in step six in the example simulation of the present invention;

[0031] Figure 6 These are the calculation results and calculation time graphs in the simulation of the embodiments of the present invention. Specific embodiments

[0032] The present invention will be described in detail below, and its features and advantages will become clearer and more definite with these descriptions.

[0033] Here, the special term "exemplary" means "serving as an example, embodiment, or illustration". Any embodiment described as "exemplary" here does not have to be construed as superior to or better than other embodiments.

[0034] The present invention provides a method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images, as Figure 1 shown, including the following steps:

[0035] Step 1: According to the DN values and calibration coefficients of the two-band long-wave infrared images obtained on orbit, obtain the spectral radiance R of the two-band long-wave infrared images.

[0036] The DN value and the spectral radiance R have the following relationship:

[0037] R = a0 + a1·DN + a2·DN 2 (1)

[0038] Wherein, a0, a1, and a2 are calibration coefficients, usually provided after ground calibration or on-board calibration; the spectral radiances of the two-band long-wave infrared images are denoted as R1 and R2, R1 is the spectral radiance of the first-band long-wave infrared image, preferably with a central wavelength of about 11 μm; R2 is the spectral radiance of the second-band long-wave infrared image, preferably with a central wavelength of about 12 μm.

[0039] Step 2: According to the spectral radiances of the two-band long-wave infrared images, use Planck's formula to determine the brightness temperatures T of the two-band long-wave infrared images.

[0040] The spectral radiance R and the brightness temperature T have the following relationship:

[0041] T = K2 / ln(K1 / R + 1) (2)

[0042] Wherein,

[0043]

[0044] h is Planck's constant, c is the speed of light, and k B is Boltzmann's constant.

[0045] The brightness temperatures of the first-band long-wave infrared image and the second-band long-wave infrared image are respectively denoted as T1 and T2.

[0046] Step 3: Determine the reflectance ρ of the visible light red spectral band image and the near-infrared image based on the DN value of the visible light image, the DN value of the near-infrared image, and the calibration coefficient. The visible light image, the near-infrared image, and the two spectral band long-wave infrared images are taken of the same area, and preprocessing processes such as geometric correction have been completed in advance. The relationship between the DN value and the reflectance ρ is as follows:

[0047] ρ = b0 + b1·DN + b2·DN 2 (3)

[0048] where b0, b1, and b2 are calibration coefficients, usually provided after ground calibration or on-orbit calibration; the reflectances of the visible light red spectral band image and the near-infrared image are denoted as ρ1 and ρ2 respectively.

[0049] Step 4: Determine the NDVI index through the reflectances of the visible light red spectral band image and the near-infrared image.

[0050] The NDVI index is determined using the following formula:

[0051]

[0052] Step 5: To reduce the computational load, based on the reflectance of the visible light red spectral band image, the NDVI index, and the vegetation coverage P v , obtain the emissivity of the two spectral band long-wave infrared images by combining the following empirical formula.

[0053]

[0054] The emissivity of the first spectral band long-wave infrared image is ε1, and the emissivity of the second spectral band long-wave infrared image is denoted as ε2.

[0055] Step 6: Determine the surface temperature based on the emissivities and the bright temperatures of the two spectral band long-wave infrared images.

[0056] T land = A0 + A1(T1 + T2) + A2(T1 - T2) (6)

[0057] where:

[0058]

[0059] The coefficients are selected as:

[0060]

[0061] In Steps 1, 2, 3, 4, 5, and 6, 16-bit unsigned integers in OpenVX are used to interact with the processing unit, and the floating-point numbers in the data processing process are converted into 16-bit fixed-point numbers with a code equivalent of 2-16 。

[0062] In Step 1, Step 2, Step 3, Step 5, and Step 6, each calculation can be split into multiple calculation parts. When there are multiple processing units, the calculation process is distributed to each processing unit, and the operation time is shortened through parallel computing. For example, in Step 6, the land surface temperatures of different pixels are calculated by multiple processing units.

[0063] Step 1 and Step 2 use two-band long-wave infrared images, and Step 3, Step 4, and Step 5 use visible light red spectral band images and near-infrared images. When there are multiple processing units, Step 1, Step 2, and Step 3, Step 4, Step 5 are parallel processing.

[0064] In Step 2, using the following approximate Planck formula, the logarithmic operation is changed to a radical operation, and a C code model based on the OpenVX API is adopted. Among them, the K part is a constant, which can be provided for on-board use after ground calculation. According to the above, the spectral radiance R is stored using 16-bit fixed-point numbers, and the square root of 16-bit fixed-point numbers is solved using the bisection method. The computational complexity is 0(Bits R ), Bits R is the number of binary digits of R.

[0065]

[0066] Where:

[0067]

[0068] In Step 6, the operation is implemented in the following way:

[0069] Calculate A1, A2, (T1 + T2), (T1 - T2). A1 and A2 are implemented using one processing unit. The emissivities of the two-band long-wave infrared images are input, and two coefficients A1 and A2 are output; (T1 + T2) and (T1 - T2) are implemented using one processing unit. The brightness temperatures of the two-band long-wave infrared images are input, and the sum of brightness temperatures and the difference in brightness temperatures are output;

[0070] Calculate A1(T1 + T2) and A2(T1 - T2). A1(T1 + T2) is implemented using one processing unit. The coefficient A1 and the sum of brightness temperatures are input, and the product is output; A2(T1 - T2) is implemented using one processing unit. The coefficient A2 and the difference in brightness temperatures are input, and the product is output;

[0071] Use one processing unit to calculate A0 + A1(T1 + T2) + A2(T1 - T2). The product values A1(T1 + T2) and A2(T1 - T2) from the previous step are input. A0 is an internal parameter of the processing unit and does not need to be input separately; the sum is output as the final result.

[0072] Inside the processing unit, the intermediate quantity of the above process is temporarily stored in 32-bit integers and processed into 16-bit fixed-point numbers for output after the calculation is completed.

[0073] The following example is used to illustrate the above method. The image used in this example is a LandSat9 image. The long-wave infrared images are from the B10 (central wavelength 10.9μm) and B11 (central wavelength 12.0μm) spectral bands of LandSat9, and the visible and near-infrared images are from the B4 (central wavelength 0.655μm) and B5 (central wavelength 0.865μm) spectral bands. The above central wavelengths are all reserved with three significant figures.

[0074] The hardware is an OBT Yulong810A chip, and the number of chips is 1 (including 8 parallel processing units). After each calculation step, the calculation result is saved and opened using ENVI software. Here, the original image is split into multiple 1024×1024 sub-blocks to fully utilize the parallel processing ability of the chip.

[0075] Step 1: Using formula (1), substitute the calibration coefficients to convert the DN values of the two spectral bands of long-wave infrared images into spectral radiance. Here, the calibration coefficients for the spectral band with a central wavelength of 10.9μm (B10) are {0.1, 3.8e-4, 0}, and the calibration coefficients for the spectral band with a central wavelength of 12μm (B11) are {0.1, 3.49e-4, 0}; the calculation results are as Figure 2 shown.

[0076] Step 2: Using the approximate Planck formula (7), convert the spectral radiance of long-wave infrared into brightness temperature. The calculation results are as Figure 3 shown.

[0077] Step 3: Using formula (3), determine the reflectance of the visible red spectral band image and the near-infrared image according to the DN values of the visible light image, the DN values of the near-infrared image, and the calibration coefficients; the calibration coefficients for the visible spectral band B4 and the near-infrared spectral band B5 are {-0.1, 2e-5, 0} and {-0.1, 2e-5, 0}, respectively.

[0078] Step 4: Using formula (4), determine the NDVI index through the reflectance of the visible red spectral band image and the near-infrared image.

[0079] Step 5: Calculate the emissivity of the long-wave infrared images of the spectral band with a central wavelength of 10.9μm (B10) and the spectral band with a central wavelength of 12μm (B11) using formula (5). The calculation results are as Figure 4 shown.

[0080] In Steps 1, 2, 3, 4, and 5, 16-bit unsigned integers in OpenVX are used to interact with the processing unit, and the floating-point numbers in the data processing process are converted into 16-bit fixed-point numbers with a code equivalent of 2. -16 .

[0081] Steps 1, 2 and Steps 3, 4, 5 are parallel processing.

[0082] Step 6: Substitute the long-wave infrared brightness temperature and the long-wave infrared reflectivity into formula (6), and the operation is specifically implemented in the following manner:

[0083] Calculate A1, A2, (T1 + T2), (T1 - T2). A1 and A2 are implemented using one processing unit. The emissivity of the long-wave infrared images in two spectral bands is input, and two coefficients A1 and A2 are output; (T1 + T2) and (T1 - T2) are implemented using one processing unit. The brightness temperature of the long-wave infrared images in two spectral bands is input, and the sum of the brightness temperatures and the difference in brightness temperatures are output;

[0084] Calculate A1(T1 + T2) and A2(T1 - T2). A1(T1 + T2) is implemented using one processing unit. The coefficient A1 and the sum of the brightness temperatures are input, and the product is output; A2(T1 - T2) is implemented using one processing unit. The coefficient A2 and the difference in brightness temperatures are input, and the product is output; The operation results of the two parts are as Figure 5 shown. The left figure is the operation result of the A1(T1 + T2) part, and the right figure is the operation result of the A2(T1 - T2) part;

[0085] Use one processing unit to calculate A0 + A1(T1 + T2) + A2(T1 - T2). The product values A1(T1 + T2) and A2(T1 - T2) from the previous step are input. A0 is an internal parameter of the processing unit and does not need to be input separately; The sum is output as the final result. The calculation result and the calculation time of a single sub-block are as Figure 6 shown.

[0086] The present invention also provides a device for on-orbit real-time inversion of surface temperature based on long-wave infrared images, including:

[0087] One or more processors;

[0088] A storage device for storing one or more programs,

[0089] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for on-orbit real-time inversion of surface temperature based on long-wave infrared images described above.

[0090] The present invention also provides a readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images described above is implemented.

[0091] The readable storage medium includes but is not limited to: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0092] The present invention also provides a computer program product. The computer program product includes: a computer program (which can also be referred to as code or instructions). When the computer program is run, the method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images described above is executed. The computer program product of the present invention is implemented using a C code model based on the OpenVX API.

[0093] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another.

[0094] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0095] The present invention has been described in detail above in combination with specific implementation manners and exemplary examples, but these descriptions should not be construed as limiting the present invention. Those skilled in the art understand that without departing from the spirit and scope of the present invention, various equivalent substitutions, modifications, or improvements can be made to the technical solution and its implementation manner of the present invention, and these all fall within the scope of the present invention. The protection scope of the present invention is subject to the appended claims.

[0096] The content not detailed in the specification of the present invention belongs to the well-known technology of those skilled in the art.

Claims

1. A method for real-time inversion of land surface temperature based on long-wave infrared images, characterized in that, Including: Obtain the spectral radiance of the two-band long-wave infrared image according to the DN value and calibration coefficient of the two-band long-wave infrared image acquired in orbit; Determine the brightness temperature of the two-band long-wave infrared image according to the spectral radiance of the two-band long-wave infrared image; Determine the reflectance of the visible light red spectral band image and the near-infrared image according to the DN value of the visible light image, the DN value of the near-infrared image and the calibration coefficient; Determine the NDVI index through the reflectance of the visible light red spectral band image and the near-infrared image; Obtain the emissivity of the two-band long-wave infrared image according to the reflectance of the visible light red spectral band image, the NDVI index and the vegetation coverage; Determine the surface temperature according to the emissivity and brightness temperature of the two-band long-wave infrared image.

2. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 1, wherein The step of obtaining the spectral radiance of the two-band long-wave infrared image according to the DN value and calibration coefficient of the two-band long-wave infrared image acquired in orbit is implemented in the following manner: There is the following relationship between the DN value and the spectral radiance: R = a0 + a1·DN + a2·DN 2 Where R is the spectral radiance, and a0, a1, a2 are calibration coefficients, which are usually provided after ground calibration or on-board calibration.

3. The method for on-orbit real-time inversion of surface temperature based on long-wave infrared images according to claim 1, wherein, The step of determining the brightness temperature of the two-band long-wave infrared image according to the spectral radiance of the two-band long-wave infrared image is implemented in the following manner: There is the following relationship between the spectral radiance and the brightness temperature: T = K2 / ln(K1 / R + 1) Where: R is the spectral radiance, T is the brightness temperature, h is the Planck constant, c is the speed of light, and k B is the Boltzmann constant.

4. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 1, characterized in that The step of determining the reflectance of the visible light red spectral band image and the near-infrared image according to the DN value of the visible light image, the DN value of the near-infrared image and the calibration coefficient is implemented in the following manner: There is the following relationship between the DN value and the reflectance: ρ = b0 + b1·DN + b2·DN 2 Where ρ is the reflectance, and b0, b1, b2 are calibration coefficients, which are usually provided after ground calibration or on-board calibration.

5. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 4, wherein The step of determining the NDVI index through the reflectance of the visible light red spectral band image and the near-infrared image is implemented in the following manner: The NDVI index is determined using the following formula: Where ρ1 and ρ2 are the reflectances of the visible light red spectral band image and the near-infrared image respectively.

6. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 5, wherein, The step of obtaining the emissivity of the two-band long-wave infrared image according to the reflectance of the visible light red spectral band image, the NDVI index and the vegetation coverage is implemented in the following manner: Where ε1 is the emissivity of the first-band long-wave infrared image, and ε2 is the emissivity of the second-band long-wave infrared image.

7. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 6, wherein The step of determining the surface temperature according to the emissivity and brightness temperature of the two-band long-wave infrared image is implemented in the following manner: T land = A0 + A1(T1 + T2) + A2(T1 - T2) Where: The coefficient is selected as: T1 and T2 are the brightness temperatures of the first-band long-wave infrared image and the second-band long-wave infrared image respectively.

8. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to any one of claims 2 to 7, characterized in that, In the steps of obtaining the spectral radiance of the two-band long-wave infrared image, determining the brightness temperature of the two-band long-wave infrared image, determining the reflectance of the visible red spectral band image and the near-infrared image, determining the NDVI index, and obtaining the emissivity of the two-band long-wave infrared image, 16-bit unsigned integers in OpenVX are used to interact with the processing unit, and the floating-point numbers in the data processing process are converted into 16-bit fixed-point numbers with a code equivalent of 2 -16 .

9. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 1, wherein In the step of determining the brightness temperature of the two-band long-wave infrared image according to the spectral radiance of the two-band long-wave infrared image, the following implementation manner is also included: Use the following approximate Planck formula to determine the brightness temperature of the two-band long-wave infrared image: The spectral radiance R is stored using 16-bit fixed-point numbers, and the square root of 16-bit fixed-point numbers is solved using the bisection method; the K part is a constant, which is calculated on the ground and provided for on-board use.

10. The method for on-orbit real-time inversion of land surface temperature based on long-wave infrared images according to claim 7, wherein The step of determining the surface temperature according to the emissivity and brightness temperature of the two-band long-wave infrared image is implemented using a parallel operation method, and the specific operation method is: Calculate A1, A2, (T1 + T2), and (T1 - T2). A1 and A2 are implemented using one processing unit. The emissivities of the two spectral band long-wave infrared images are input, and the two coefficients A1 and A2 are output. (T1 + T2) and (T1 - T2) are implemented using one processing unit. The brightness temperatures of the two spectral band long-wave infrared images are input, and the sum and difference of the brightness temperatures are output. Calculate A1(T1 + T2) and A2(T1 - T2). A1(T1 + T2) is implemented using one processing unit. The coefficient A1 and the sum of the brightness temperatures are input, and the product is output. A2(T1 - T2) is implemented using one processing unit. The coefficient A2 and the difference of the brightness temperatures are input, and the product is output. Use one processing unit to calculate A0 + A1(T1 + T2) + A2(T1 - T2). The product values A1(T1 + T2) and A2(T1 - T2) from the previous step are input. A0 is a built-in parameter of the processing unit, and the sum is output as the final result.

Images