Artificial intelligence-based remote sensing instrument radiation key component identification method and device

Abstract

The application discloses a kind of based on artificial intelligence's remote sensing instrument radiation key component identification method and device.The method includes: obtaining the radiation calibration coefficient of each fitting order of the infrared remote sensing instrument to be identified, and obtaining the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiation calibration coefficient is obtained;Respectively based on the radiation calibration coefficient of each fitting order and the temperature of each optical path component, training is carried out, and the artificial intelligence network corresponding to each fitting order is obtained;Respectively based on the artificial intelligence network corresponding to each fitting order, the rate of change of radiation calibration coefficient with the temperature of optical path component is obtained;Based on the rate of change of radiation calibration coefficient with the temperature of optical path component, key components in each optical path component are identified.The remote sensing instrument radiation key component identification method and device based on artificial intelligence disclosed in the application can accurately identify the radiation key point in the infrared remote sensing instrument by quantitatively analyzing the influence of different components on the radiation stability of the infrared remote sensing instrument.

Description

Technical Field

[0001] This invention relates to the field of remote sensing technology, and in particular to a method and apparatus for identifying key radiation components of remote sensing instruments based on artificial intelligence. Background Technology

[0002] For high-precision infrared remote sensing measurements, the radiation stability of infrared remote sensing instruments is crucial. The radiation calibration coefficient of an infrared remote sensing instrument is related to the temperature field, but currently, there is a lack of scientifically sound calculation methods to clearly define the contribution of each component of the instrument to radiation stability. Therefore, currently, key components are typically identified based on the designer's experience to ensure the radiation stability of the infrared remote sensing instrument, such as implementing temperature control facilities on critical radiation components.

[0003] However, due to the inability to quantitatively analyze the impact of different components on the radiation stability of infrared remote sensing instruments, there is insufficient temperature control and / or temperature measurement capabilities for some components with sufficiently large "influence," and / or excessive temperature control and temperature measurement capabilities for some components with less "influence." Insufficient temperature control and / or temperature measurement capabilities in some of these components with sufficiently large "influence" will lead to insufficient radiation stability and decreased accuracy of radiation measurements in infrared remote sensing instruments. Excessive temperature control and / or temperature measurement capabilities in these components with less "influence" will result in excessive overall resource consumption and increased development difficulty for satellites carrying infrared remote sensing instruments, thereby reducing the satellite's capabilities.

[0004] Therefore, accurately identifying the key components in infrared remote sensing instruments that play a crucial role in radiation stability has become a pressing technical issue that needs to be addressed in this field.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method and apparatus for identifying key radiation components of remote sensing instruments based on artificial intelligence.

[0007] To achieve the above objectives, the present invention provides a method for identifying key radiation components of remote sensing instruments based on artificial intelligence, comprising:

[0008] Obtain the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and obtain the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are obtained.

[0009] The artificial intelligence network corresponding to each fitting order is obtained by training based on the radiation calibration coefficients and the temperature of each optical path component for each fitting order.

[0010] Based on the artificial intelligence network corresponding to each fitting order, the rate of change of the radiation calibration coefficient with the temperature of each optical path component is obtained.

[0011] Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, the key components in each optical path component are identified.

[0012] In one embodiment of the present invention, obtaining the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order includes:

[0013] For each fitting order of the artificial intelligence network, the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component is obtained.

[0014] Based on the total error of the output of the artificial intelligence network corresponding to each fitting order, the partial derivatives of the radiation calibration coefficients with respect to the ambient temperature field are obtained.

[0015] Based on the partial derivative of the total error of the AI ​​network output corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field, the rate of change of the radiation calibration coefficient with respect to the temperature of each optical path component is obtained.

[0016] In one embodiment of the present invention, identifying key components in each optical path component based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component includes:

[0017] Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, the relationship between the rate of change of the temperature of each optical path component and the observed radiation of the infrared remote sensing instrument is obtained respectively.

[0018] Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, key components in each optical path component are identified.

[0019] In one embodiment of the present invention, identifying key components in each optical path component based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument includes:

[0020] Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument is determined.

[0021] The optical path components whose influence is greater than a preset value are identified as the key components, or the key components are identified based on the order of the magnitude of the influence.

[0022] The present invention also provides an artificial intelligence-based device for identifying key radiation components of remote sensing instruments, comprising:

[0023] The first acquisition module is used to acquire the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and to acquire the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are acquired.

[0024] The training module is used to train the artificial intelligence network corresponding to each fitting order based on the radiation calibration coefficients and the temperature of each optical path component for each fitting order.

[0025] The second acquisition module is used to acquire the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order.

[0026] An identification module is used to identify key components in each optical path component based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component.

[0027] In one embodiment of the present invention, the second acquisition module includes:

[0028] The first acquisition unit is used for the partial derivative of the total error of the output of the corresponding artificial intelligence network with respect to the temperature of each optical path component;

[0029] The second acquisition unit is used to acquire the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field based on the total error of the output of the artificial intelligence network corresponding to each fitting order.

[0030] The third acquisition unit is used to acquire the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field.

[0031] In one embodiment of the present invention, the identification module includes:

[0032] The fourth acquisition unit is used to acquire the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, based on the rate of temperature change of the radiation calibration coefficient with respect to the temperature of each optical path component.

[0033] The identification unit is used to identify key components in each optical path component based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument.

[0034] In one embodiment of the present invention, the identification unit is specifically used for:

[0035] Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument is determined.

[0036] The optical path components whose influence is greater than a preset value are identified as the key components, or the key components are identified based on the order of the magnitude of the influence.

[0037] In one embodiment of the present invention, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the artificial intelligence-based remote sensing instrument radiation key component identification method described above.

[0038] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the artificial intelligence-based remote sensing instrument radiation key component identification method as described above.

[0039] In one embodiment of the present invention, a computer program product includes a computer program that, when executed by a processor, implements the steps of the artificial intelligence-based remote sensing instrument radiation key component identification method as described above.

[0040] Compared with existing technologies, the present invention provides a method and apparatus for identifying key radiation components of remote sensing instruments based on artificial intelligence. By quantitatively analyzing the impact of different components on the radiation stability of infrared remote sensing instruments, it can accurately identify key components in infrared remote sensing instruments that play a crucial role in radiation stability. Attached Figure Description

[0041] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their descriptions, serve to explain the invention and do not constitute an undue limitation thereof. The realization of the invention's objectives, functional features, and advantages will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0042] Figure 1 This is a flowchart illustrating a method for identifying key radiation components of a remote sensing instrument based on artificial intelligence, according to an embodiment of the present invention.

[0043] Figure 2 This is a schematic diagram of the structure of a remote sensing instrument radiation key component identification device based on artificial intelligence according to an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0045] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings, but it should be understood that the scope of protection of the present invention is not limited to the specific embodiments.

[0046] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0047] Figure 2 This is a flowchart illustrating a method for identifying key radiation components of a remote sensing instrument based on artificial intelligence, according to an embodiment of the present invention. Figure 2 As shown, a method for identifying key radiation components of a remote sensing instrument based on artificial intelligence according to a preferred embodiment of the present invention may include steps 101, 102, 103 and 104.

[0048] To facilitate understanding of the embodiments of this application, the physical theories involved in the embodiments of this application will be explained below.

[0049] According to infrared semiconductor physics theory, the radiation response capability of infrared remote sensing instruments is greatly affected by the instrument's temperature conditions. The relationship between the observed radiation of an infrared remote sensing instrument and its digital output can usually be expressed by formula (1).

[0050]

[0051] Where R represents the observed radiance of the infrared remote sensing instrument; DN represents the digital output of the infrared remote sensing instrument; {a i} represents the radiometric calibration coefficient corresponding to the fitting order i, used to indicate the conversion relationship between the digital output DN of the infrared remote sensing instrument and the observed radiance R; N represents the highest order of fitting; DN i In this context, 'i' does not represent exponentiation, but rather the fitting order 'i' corresponding to DN.

[0052] The radiometric calibration factor is typically a parameter value under a fixed operating condition, and it will change accordingly as the operating condition changes. This operating condition can include temperature conditions. Temperature conditions can usually be represented by the temperature field {T} of components along the radiation path of an infrared remote sensing instrument.j The expression is represented by |j=1,2,...,K}. Here, K is the number of temperature measurement points, which is the number of optical components in the radiation optical path.

[0053] The radiometric stability of an infrared remote sensing instrument can be represented by the stability of its radiometric calibration coefficient. That is, when the actual radiometric calibration coefficient of the infrared remote sensing instrument changes to {a...} i When '}, because infrared remote sensing instruments cannot be calibrated constantly, but require a period of time to perform a calibration operation to obtain new radiometric calibration coefficients, the radiometric calibration coefficients {a} obtained at the time of the previous calibration operation are still used when calculating the digital output DN to the observed radiance R. i}. In {a i Under conditions of small change, due to {a i}≈{a i This ensures the accuracy of remote sensing radiometric measurements.

[0054] However, when the radiometric response of an infrared remote sensing instrument varies significantly, the radiometric calibration coefficient {a} obtained from the most recent calibration operation is still used when calculating the digital output DN to the observed radiance R. i}, {a i} and the true radiation calibration coefficient {a i The difference between '} is relatively large. Let Δ = {a} i}-{a i When Δ exists and is a large quantity, the difference between the observed radiation R obtained by conversion and the actual measured value R′ is also large.

[0055] Step 101: Obtain the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and obtain the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are obtained.

[0056] In practice, the infrared remote sensing instrument to be identified can be an infrared remote sensing instrument carried by a satellite or drone. The radiometric calibration coefficient {a} of the infrared remote sensing instrument to be identified can be obtained first, based on (but not limited to) the vacuum calibration experiment results before launch. i |i=0,1,2,...,N}, and record the working status of the infrared remote sensing instrument at the moment when the above radiometric calibration coefficients are obtained.

[0057] The highest order N in the fit indicates how many polynomials were used to fit the radiometric calibration curve. Generally, if the response linearity of the infrared remote sensing instrument is high, N can be 1 or 2; however, the embodiments of this application do not limit the specific value of N.

[0058] In some feasible implementations, the aforementioned vacuum calibration experimental results may be obtained based on laboratory vacuum chamber calibration or other radiation calibration methods.

[0059] In some feasible implementations, the operating status of the aforementioned infrared remote sensing instrument can be determined by the physical temperature {T} of each optical path component within the infrared remote sensing instrument. j |j=1,2...,K} serves as an indicator of the state. The set of temperatures for all optical path components is called the temperature condition. The aforementioned optical path components are those whose temperatures are measurable and which are used in the optical path.

[0060] Step 102: Train the network based on the radiation calibration coefficients and the temperature of each optical path component for each fitting order to obtain the artificial intelligence network corresponding to each fitting order.

[0061] In practical implementation, a mapping relationship from the temperature conditions of the infrared remote sensing instrument to the radiometric calibration coefficients can be trained based on multilayer perceptron (MLP) neural networks or other artificial intelligence (AI) methods or models, according to the radiometric calibration coefficients for each fitting order and the temperature of each optical path component. This mapping relationship can be expressed as follows: in, This represents a mapping network obtained by training using artificial intelligence methods, showing the relationship between the temperature conditions of the infrared remote sensing instrument to be identified and its radiometric calibration coefficients. This mapping network is an artificial intelligence network.

[0062] in, In this context, 'i' indicates that the artificial intelligence network needs to target the radiation calibration coefficient set {a}. i Each radiation calibration factor a in} i The training process involves training (N+1) artificial intelligence networks in step 102, where the calibration uses an Nth-order curve fitting radiation calibration equation as shown in the aforementioned formula (1).

[0063] Step 103: Based on the artificial intelligence network corresponding to each fitting order, obtain the rate of change of the radiation calibration coefficient with the temperature of each optical path component.

[0064] In practice, since the artificial intelligence network corresponding to each fitting order can be used to indicate the mapping relationship between the temperature of the optical path element and the radiation calibration coefficient of that fitting order, the rate at which the radiation calibration coefficient changes with the temperature of each optical path component can be obtained by combining the artificial intelligence networks corresponding to each fitting order.

[0065] In some feasible implementations, the rate of change of the radiation calibration coefficients with the temperature of each optical path component is obtained based on the artificial intelligence network corresponding to each fitting order, including:

[0066] For each fitting order, the partial derivative of the total error of the AI ​​network output with respect to the temperature of each optical path component is obtained.

[0067] Based on the total error of the output of the artificial intelligence network corresponding to each fitting order, the partial derivatives of the radiation calibration coefficients with respect to the ambient temperature field are obtained.

[0068] Based on the partial derivative of the total error of the AI ​​network output corresponding to each fitting order with respect to the temperature of each optical component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field, the rate of change of the radiation calibration coefficient with respect to the temperature of each optical component is obtained.

[0069] In some feasible implementations, for each artificial intelligence network The artificial intelligence network can be obtained through mathematical methods. The total error E of the output relative to the input {T} j The Jacobian matrix of}, i.e., the total error E for each input {T} j The partial derivatives of |j=1,2,...,K} yield The Jacobian matrix described above can represent the sensitivity of the total error E to temperature changes in the optical components of the infrared remote sensing instrument to be identified.

[0070] In some feasible implementations, computational artificial intelligence networks (For convenience, i) is omitted here. The Jacobian matrix of the artificial intelligence network... The total error E of the output is relative to each input {T} j The partial derivatives of |j = 1, 2, ..., K}. This is illustrated using artificial intelligence networks. Taking a back propagation neural network (BP neural network) as an example, the Jacobian matrix is ​​expressed as follows:

[0071]

[0072] Where E represents an artificial intelligence network The total error of the output; the variables on the right side of the equals sign are all trained artificial intelligence networks. The network parameters, including weights and biases.

[0073] It should be noted that the above formula (2) is based on artificial intelligence networks. Using a backpropagation (BP) neural network as an example, but the embodiments of this application do not limit the use of artificial intelligence networks. A backpropagation (BP) network must be used; the above is merely a formula example. Then, based on this artificial intelligence network... The formula for the total error E of the output is used to calculate the result with {a}. i The artificial intelligence network represented by} The total error E of the output is relative to each input {T} j Partial derivatives of |j=1,2...,K}

[0074] In some feasible implementations, an artificial intelligence network can be defined first. (For convenience, i is omitted here) The expression for the total error E of the output is obtained, and the derivative is taken with respect to the ambient temperature field to obtain the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field.

[0075] Artificial Intelligence Network The total error E of the output can be expressed as:

[0076]

[0077] Where z represents an artificial intelligence network The output (i.e., the artificial intelligence network) The predicted radiation calibration coefficients are used to train the artificial intelligence network. The label is distinguished by the symbol z); a is the radiation calibration coefficient obtained by pre-calibration in a vacuum chamber or other methods (used for training artificial intelligence networks). (tags).

[0078] Formula (3) for input T j Differentiation yields

[0079]

[0080] It should be noted that the above formula (3) is only for artificial intelligence networks. One of the forms of the total error E output is... In the embodiments of this application, the artificial intelligence network... The total error E of the output can also be expressed using other formulas, including equivalent transformations of formula (3), such as removing or modifying the coefficients. wait.

[0081] In some feasible implementations, the aforementioned results can be combined to calculate the trained artificial intelligence network. The network parameters are used to represent symbols.

[0082] In some feasible implementations, formulas (2) and (4) can be combined to calculate the result.

[0083]

[0084] Understandable, This is the rate at which the radiation calibration coefficient changes with the temperature of the optical path components.

[0085] Step 104: Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, identify the key components in each optical path component.

[0086] In actual implementation, it can be Substituting into the radiation calibration formula, i.e., the aforementioned formula (1), calculate... exist Based on this, the key radiation points of the infrared remote sensing instrument to be identified can be analyzed. These key points are critical components, specifically those where temperature changes significantly affect the radiation calibration coefficient. Generally, The larger the point, the more critical it is. The larger the optical path component, the more critical it is.

[0087] In some feasible implementations, preset values ​​can be pre-set, and the determination can be made for each optical path component (corresponding to each value of j). Is it greater than a preset value? If it is greater than or equal to the preset value, then the optical path component is a critical component; if it is less than the preset value, then the optical path component is not a critical component. In some feasible implementations, each optical path component can be assigned a corresponding value. Sort the values ​​in descending order and select the M largest ones. The corresponding optical path components are considered key components. Here, M is less than K. For example, M can be 1, 2, or 3, etc.

[0088] It should be noted that the key radiation points (i.e., key components) of an infrared remote sensing instrument are crucial to the stability of its radiation response. Therefore, after analyzing and identifying these key radiation points, a series of key measures, such as strengthening temperature control and / or strengthening temperature measurement, can be taken to improve the stability of the infrared remote sensing instrument. This stability refers to the stability of the radiation values ​​obtained by the infrared remote sensing instrument through measurement.

[0089] In some feasible implementations, key components in each optical path are identified based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, including:

[0090] Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, the relationship between the rate of change of the temperature of each optical path component and the observed radiation of the infrared remote sensing instrument is obtained.

[0091] Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the key components in each optical path component are identified.

[0092] In some feasible implementations, based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, key components in each optical path component are identified, including: determining the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument.

[0093] Optical path components whose impact exceeds a preset value are identified as critical components, or critical components are identified based on the order of their impact.

[0094] In actual implementation, formulas (2) and (1) can be combined for calculation. get

[0095]

[0096] By following the steps above, the dynamic range of the entire domain can be obtained. curve. It can be used to indicate the degree to which temperature changes in optical path components affect the amount of radiation observed by infrared remote sensing instruments.

[0097] In some feasible implementations, for The curves represent the digital output (DN) of the infrared remote sensing instrument on the horizontal axis (usually the x-axis) and the rate of change of the observed radiance of the infrared remote sensing instrument with respect to the temperature of a certain optical path component on the vertical axis (usually the y-axis). Based on K curves, the degree of influence of the temperature changes of K optical path components on the observed radiance of the infrared remote sensing instrument can be determined.

[0098] Understandably, for a specific infrared remote sensing instrument, in order to ensure its radiation stability, it is desirable for the instrument to be as insensitive to temperature as possible during operation. The smaller the better. Therefore, this application can quantitatively calculate the temperature sensitivity of different optical path components in an infrared remote sensing instrument. Based on the sensitivity analysis, the radiation key points of the infrared remote sensing instrument can be obtained. Furthermore, the radiation stability of the infrared remote sensing instrument can be improved by strengthening the temperature control of the above-mentioned radiation key points and other methods to ensure temperature stability.

[0099] To facilitate understanding of the above embodiments of this application, an implementation process of an artificial intelligence-based method for identifying key radiation components of remote sensing instruments is described below.

[0100] Step S1, based on (but not limited to) the pre-launch laboratory vacuum chamber radiometric calibration experiment of the infrared remote sensing instrument, obtain the radiometric calibration coefficient {a} of the infrared remote sensing instrument. i|i=0,1,2,...,N}, and record the temperature conditions {T} of the infrared remote sensing instrument at the moment when the radiometric calibration coefficient is obtained. k |k=1,2,...,K}。 Where N represents the highest order of the fit; K is the number of measurable and usable temperatures on the optical path components, i.e., the number of the aforementioned optical path elements.

[0101] Step S2: Based on artificial intelligence methods, train an AI network to indicate the mapping relationship from the temperature conditions of the infrared remote sensing instrument to be identified to the radiometric calibration coefficients. Step S3, Calculate the artificial intelligence network (For convenience, i) is omitted here. The Jacobian matrix of the artificial intelligence network... The total error E of the output is relative to each input {T} j Partial derivatives of |j=1,2...,K} Step S4, Define the artificial intelligence network The expression for the total error E of the output is obtained by differentiating it with respect to the ambient temperature field, thus yielding the result from the artificial intelligence network. The total error E of the output is the partial derivative of the radiation scaling coefficients with respect to the ambient temperature field. Step S5: Combining the results of steps S3 and S4, calculate the trained artificial intelligence network. The network parameters are used to represent symbols. Step S6: Calculate the partial derivative of the observed radiance of the infrared remote sensing instrument to be identified with respect to the temperature of the optical path components using formula (1). Thus, the dynamic range of the entire domain is obtained. The curves, based on which the temperature changes of K optical path components in the optical path can be determined, can determine the degree of influence of the temperature changes of K optical path components on the observed radiation of the infrared remote sensing instrument. Thus, the key components among the K optical path components can be identified based on the degree of influence of the temperature changes of K optical path components on the observed radiation of the infrared remote sensing instrument.

[0102] The beneficial effect of this invention is that, by quantitatively analyzing the influence of different components on the radiation stability of infrared remote sensing instruments, it is possible to accurately identify the key components in infrared remote sensing instruments that play a crucial role in radiation stability.

[0103] The following describes the AI-based remote sensing instrument radiation key component identification device provided by the present invention. The AI-based remote sensing instrument radiation key component identification device described below and the AI-based remote sensing instrument radiation key component identification method described above can be referred to in correspondence.

[0104] Figure 2This is a schematic diagram of a remote sensing instrument radiation key component identification device based on artificial intelligence according to an embodiment of the present invention. Based on the content of any of the above embodiments, such as... Figure 2 As shown, the device includes a first acquisition module 201, a training module 202, a second acquisition module 203, and a recognition module 204, wherein:

[0105] The first acquisition module 201 is used to acquire the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and to acquire the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are acquired; the training module 202 is used to train based on the radiometric calibration coefficients of each fitting order and the temperature of each optical path component respectively, and to acquire the artificial intelligence network corresponding to each fitting order.

[0106] The second acquisition module 203 is used to acquire the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order.

[0107] The identification module 204 is used to identify key components in each optical path component based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component.

[0108] In some feasible implementations, the second acquisition module 203 includes:

[0109] The first acquisition unit is used for the partial derivative of the total error of the corresponding artificial intelligence network output with respect to the temperature of each optical path component.

[0110] The second acquisition unit is used to obtain the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field based on the total error of the output of the artificial intelligence network corresponding to each fitting order.

[0111] The third acquisition unit is used to acquire the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field.

[0112] In some feasible implementations, the identification module 204 includes:

[0113] The fourth acquisition unit is used to acquire the relationship between the rate of temperature change of each optical path component and the observed radiation amount of the infrared remote sensing instrument, based on the rate of temperature change of the radiation calibration coefficient with respect to the rate of temperature change of each optical path component.

[0114] The identification unit is used to identify key components in each optical path component based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument.

[0115] In some feasible implementations, the identification unit is specifically used for:

[0116] Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument is determined.

[0117] Optical path components whose impact exceeds a preset value are identified as critical components, or critical components are identified based on the order of their impact.

[0118] The artificial intelligence-based remote sensing instrument radiation key component identification device provided in this embodiment of the invention is used to execute the artificial intelligence-based remote sensing instrument radiation key component identification method of the present invention. Its implementation method is consistent with the implementation method of the artificial intelligence-based remote sensing instrument radiation key component identification method provided in this invention, and can achieve the same beneficial effects, so it will not be described again here.

[0119] This AI-based remote sensing instrument radiation critical component identification device is used in the AI-based remote sensing instrument radiation critical component identification methods of the foregoing embodiments. Therefore, the descriptions and definitions in the AI-based remote sensing instrument radiation critical component identification methods of the foregoing embodiments can be used for understanding the various execution modules in the embodiments of the present invention.

[0120] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute an artificial intelligence-based method for identifying key radiation components of a remote sensing instrument. This method includes: acquiring radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and acquiring the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are acquired; training the device based on the radiometric calibration coefficients of each fitting order and the temperature of each optical path component to acquire an artificial intelligence network corresponding to each fitting order; acquiring the rate of change of the radiometric calibration coefficients with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order; and identifying the observed radiation of key components in each optical path component based on the rate of change of the radiometric calibration coefficients with the temperature of each optical path component.

[0121] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0122] The processor 310 in the electronic device provided in this embodiment of the invention can call the logical instructions in the memory 330. Its implementation method is consistent with the implementation method of the remote sensing instrument radiation key component identification method based on artificial intelligence provided in this invention, and can achieve the same beneficial effect. It will not be described again here.

[0123] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, and when the program instructions are executed by the computer, the computer is able to execute the artificial intelligence-based method for identifying key radiation components of remote sensing instruments provided by the above methods. The method includes: acquiring radiometric calibration coefficients of multiple fitting orders of an infrared remote sensing instrument to be identified, and acquiring the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are acquired; training based on the radiometric calibration coefficients of each fitting order and the temperature of each optical path component to acquire an artificial intelligence network corresponding to each fitting order; acquiring the rate of change of the radiometric calibration coefficients with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order; and identifying the observed radiation of key components in each optical path component based on the rate of change of the radiometric calibration coefficients with the temperature of each optical path component.

[0124] When the computer program product provided in this embodiment of the invention is executed, it implements the above-mentioned method for identifying key radiation components of remote sensing instruments based on artificial intelligence. Its specific implementation method is consistent with the implementation method described in the aforementioned method embodiment, and can achieve the same beneficial effects, which will not be repeated here.

[0125] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program performs the aforementioned artificial intelligence-based remote sensing instrument radiation key component identification method. The method includes: acquiring radiometric calibration coefficients of multiple fitting orders of an infrared remote sensing instrument to be identified, and acquiring the temperature of each optical path component in the infrared remote sensing instrument at the time the radiometric calibration coefficients are acquired; training based on the radiometric calibration coefficients of each fitting order and the temperature of each optical path component to acquire an artificial intelligence network corresponding to each fitting order; acquiring the rate of change of the radiometric calibration coefficients with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order; and identifying the observed radiation of key components in each optical path component based on the rate of change of the radiometric calibration coefficients with the temperature of each optical path component.

[0126] When the computer program stored on the non-transitory computer-readable storage medium provided in this embodiment of the invention is executed, it implements the above-mentioned method for identifying key radiation components of remote sensing instruments based on artificial intelligence. Its specific implementation method is consistent with the implementation method described in the aforementioned method embodiment, and can achieve the same beneficial effects, so it will not be repeated here.

[0127] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0128] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0129] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0130] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes. The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

[0131] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be found in the description of the method embodiments.

[0132] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0133] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for identifying key radiation components of remote sensing instruments based on artificial intelligence, characterized in that, include: Obtain the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and obtain the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are obtained. The artificial intelligence network corresponding to each fitting order is obtained by training based on the radiation calibration coefficients and the temperature of each optical path component for each fitting order. Based on the artificial intelligence network corresponding to each fitting order, the rate of change of the radiation calibration coefficient with the temperature of each optical path component is obtained. Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, the key components in each optical path component are identified; The step of obtaining the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the artificial intelligence network corresponding to each fitting order includes: For each fitting order of the artificial intelligence network, the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component is obtained. Based on the total error of the output of the artificial intelligence network corresponding to each fitting order, the partial derivatives of the radiation calibration coefficients with respect to the ambient temperature field are obtained. Based on the partial derivative of the total error of the AI ​​network output corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field, the rate of change of the radiation calibration coefficient with respect to the temperature of each optical path component is obtained.

2. The method for identifying key radiation components of remote sensing instruments based on artificial intelligence according to claim 1, characterized in that, The method of identifying key components in each optical path component based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component includes: Based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component, the relationship between the rate of change of the temperature of each optical path component and the observed radiation of the infrared remote sensing instrument is obtained respectively. Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, key components in each optical path component are identified.

3. The method for identifying key radiation components of remote sensing instruments based on artificial intelligence according to claim 2, characterized in that, The method of identifying key components in each optical path component based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument includes: Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument is determined. The optical path components whose influence is greater than a preset value are identified as the key components, or the key components are identified based on the order of the magnitude of the influence.

4. A remote sensing instrument radiation key component identification device according to the artificial intelligence-based remote sensing instrument radiation key component identification method as described in any one of claims 1-3, characterized in that, include: The first acquisition module is used to acquire the radiometric calibration coefficients of multiple fitting orders of the infrared remote sensing instrument to be identified, and to acquire the temperature of each optical path component in the infrared remote sensing instrument at the time when the radiometric calibration coefficients are acquired. The training module is used to train the artificial intelligence network corresponding to each fitting order based on the radiation calibration coefficients and the temperature of each optical path component for each fitting order. The second acquisition module is used to acquire, based on the artificial intelligence network corresponding to each fitting order, the rate of change of the radiation calibration coefficient with the temperature of each optical path component, including: For each fitting order of the artificial intelligence network, the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component is obtained. Based on the total error of the output of the artificial intelligence network corresponding to each fitting order, the partial derivatives of the radiation calibration coefficients with respect to the ambient temperature field are obtained. Based on the partial derivative of the total error of the AI ​​network output corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field, the rate of change of the radiation calibration coefficient with respect to the temperature of each optical path component is obtained. An identification module is used to identify key components in each optical path component based on the rate of change of the radiation calibration coefficient with the temperature of each optical path component.

5. The remote sensing instrument radiation key component identification device based on artificial intelligence according to claim 4, characterized in that, The second acquisition module includes: The first acquisition unit is used to acquire, for each fitting order corresponding to the artificial intelligence network, the partial derivative of the total error of the output of the artificial intelligence network with respect to the temperature of each optical path component. The second acquisition unit is used to acquire the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field based on the total error of the output of the artificial intelligence network corresponding to each fitting order. The third acquisition unit is used to acquire the rate of change of the radiation calibration coefficient with the temperature of each optical path component based on the partial derivative of the total error of the output of the artificial intelligence network corresponding to each fitting order with respect to the temperature of each optical path component and the partial derivative of the radiation calibration coefficient with respect to the ambient temperature field.

6. The artificial intelligence-based remote sensing instrument radiation key component identification device according to claim 4 or 5, characterized in that, The identification module includes: The fourth acquisition unit is used to acquire the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, based on the rate of temperature change of the radiation calibration coefficient with respect to the temperature of each optical path component. The identification unit is used to identify key components in each optical path component based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument.

7. The remote sensing instrument radiation key component identification device based on artificial intelligence according to claim 6, characterized in that, The identification unit is specifically used for: Based on the relationship between the rate of temperature change of each optical path component and the observed radiation of the infrared remote sensing instrument, the degree of influence of the temperature change of each optical path component on the observed radiation of the infrared remote sensing instrument is determined. The optical path components whose influence is greater than a preset value are identified as the key components, or the key components are identified based on the order of the magnitude of the influence.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for identifying key radiation components of remote sensing instruments based on artificial intelligence as described in any one of claims 1 to 3.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for identifying key radiation components of remote sensing instruments based on artificial intelligence as described in any one of claims 1 to 3.

Images