Infrared thermography method for automatic calibration of sensor parameters based on ambient temperature

Abstract

This invention relates to the field of infrared thermal imaging technology, and particularly to an infrared thermal imaging method for automatically calibrating sensor parameters based on ambient temperature. The method includes: Step 1: Acquiring temperature data from a built-in reference source and determining whether temperature inaccuracy has occurred; if yes, proceeding to Step 2; otherwise, proceeding to Step 4; Step 2: Calculating the failure probability of the target temperature sensor and determining whether the temperature sensor is faulty; if yes, proceeding to Step 3; otherwise, proceeding to Step 4; Step 3: Fusing the sensor failure probability to optimize the inaccuracy judgment of the built-in reference source and predicting the temperature of the built-in reference source; then, constructing a temperature compensation model and using ambient temperature and historical data to correct the output thermal image data; Step 4: Determining whether the ambient temperature exceeds the upper limit of the predicted operating temperature; if yes, constructing a high-temperature adaptive compensation model for nonlinear compensation; otherwise, returning to Step 1.

Description

Technical Field

[0001] This invention relates to the field of infrared thermal imaging technology, and in particular to an infrared thermal imaging method based on automatic calibration of sensor parameters according to ambient temperature. Background Technology

[0002] Calibration of infrared thermal imaging equipment is a crucial step in ensuring the accuracy of its temperature measurements. Automatic calibration of sensor parameters based on ambient temperature is essential to compensate for measurement errors caused by temperature drift within the thermal imager itself. Its core principle is closed-loop compensation based on a built-in reference source and temperature sensor.

[0003] Common thermal imagers integrate a miniature reference source (e.g., a temperature-controlled miniature blackboard, a thermally stabilizing element) with a highly stable temperature and a known emissivity (typically close to 1). The temperature of this reference source is monitored in real time by a high-precision temperature sensor (e.g., a Pt100 platinum resistance thermometer).

[0004] The long-term stability of the built-in miniature reference source is crucial; if its temperature is inaccurate, the entire basis of automatic calibration is flawed. Therefore, we propose a method to effectively determine if the miniature reference source is malfunctioning and, if so, to perform correction to ensure the accuracy of temperature measurements. Summary of the Invention

[0005] This invention monitors the stability of the built-in reference source and the status of its associated temperature sensor, incorporating the probability of temperature sensor failure into the framework for judging the inaccuracy of the built-in reference source temperature; and after the temperature sensor fails, it uses ambient temperature data and historical data for adaptive temperature compensation to ensure the accuracy of temperature measurement.

[0006] The technical solution proposed in this invention is: an infrared thermal imaging method based on automatic calibration of sensor parameters according to ambient temperature, the method comprising: An infrared thermal imaging method based on automatic calibration of sensor parameters according to ambient temperature, characterized in that the method includes: Step 1: Obtain temperature data from the built-in reference source to analyze its stability and determine if temperature inaccuracy has occurred; if so, proceed to Step 2; otherwise, proceed to Step 4. Step 2: Calculate the failure probability of the target temperature sensor and determine whether the temperature sensor is faulty. If it is, proceed to Step 3; otherwise, proceed to Step 4. Step 3: Fuse sensor failure probabilities to optimize the judgment of inaccuracy of the built-in reference source and predict the temperature of the built-in reference source; then, build a temperature compensation model and use ambient temperature and historical data to correct the output thermal image data. Step 4: Determine whether the ambient temperature exceeds the upper limit of the predicted operating temperature; if so, construct a high-temperature adaptive compensation model and perform nonlinear compensation to ensure the temperature measurement accuracy under high-temperature conditions; otherwise, return to step 1.

[0007] Preferably, the step of acquiring temperature data from the built-in reference source for analyzing the stability of the built-in reference source and determining whether temperature inaccuracy has occurred includes: Determine the short-term stability and long-term temperature drift of the built-in reference source to determine if temperature inaccuracy has occurred. Use a time period Long-term temperature drift data were used to construct a long-term drift time series. Based on the long-term drift time series, a pre-trained trend analysis model is used to predict the future temperature inaccuracy time; the trend analysis model is an autoregressive integral moving average (ARIMA) model.

[0008] Preferably, determining the short-term stability and long-term temperature drift of the built-in reference source includes: Obtain temperature data from the built-in reference source within a preset time window and construct a temperature sequence from the built-in reference source. ;in, Indicates the length of the time window; Indicates the first time within the time window One built-in reference source temperature; The acquired temperature data from the built-in reference source is compared with the preset temperature threshold. Compare and calculate the temperature deviation within the corresponding time window. , ; Calculating short-term stability includes: Calculate the standard deviation of temperature deviation within the time window. ,if If the short-term stability of the built-in reference source is deemed abnormal, an alarm will be triggered; where, Indicates the standard deviation threshold; Calculate the long-term temperature drift of the built-in reference source, including: Sliding time window, calculation The moving average deviation of temperature deviation over a continuous time window, i.e., the long-term temperature drift. ;in, Indicates the total number of time windows; if If so, the temperature judgment is inaccurate; among them, Temperature deviation threshold.

[0009] Preferably, the step of using a pre-trained trend analysis model to predict future temperature inaccuracy time includes: Obtaining the time series of long-term drift , ;in, Indicates the temperature sampling frequency within the time window; right Perform differential processing to obtain the differential time series. ; Fitting the difference time series: ;in, Indicates the order of autoregression. Indicates the order of the moving average. Represents white noise. Indicates the intercept; , Indicates the weighting coefficient; Then, The data is input into a pre-trained trend analysis model; the prediction time is then determined. Long-term temperature drift ; If, | Then determine the time. The temperature was inaccurate.

[0010] Preferably, the calculation of the failure probability of the target temperature sensor and the determination of whether the temperature sensor is faulty include: The temperature sensor status is predicted and the failure probability is calculated using Kalman filter residual analysis, including: Constructing the state equations: ; ;in, Indicates the target sensor at the current time. The measured temperature value; Indicates the target temperature sensor The temperature value measured at the previous moment; Indicates temperature noise; Indicates observation noise; Using the Kalman filter algorithm, the temperature value at the next moment is predicted based on the sensor's current temperature value. ; Calculate residuals ; The probability of target sensor failure ;in, This represents the standard deviation of the temperature residual under normal operating conditions. if If the target temperature sensor is faulty, then the fault is determined; where, This represents the failure probability threshold.

[0011] Preferably, the fusion sensor failure probability optimizes the judgment of the inaccuracy of the built-in reference source and predicts the temperature of the built-in reference source, including: Obtain the infrared radiation value of the built-in reference source, and calculate the deviation between the theoretical radiation value and the obtained infrared radiation value of the built-in reference source. ;in, This indicates the radiation value of the built-in reference source. Indicates the theoretical radiation value; ;in, This indicates the temperature of the blackbody. Spectral radiance under; Indicates the emissivity of the built-in reference source; , Indicates the reference source detector gain and offset; The deviation in infrared radiation values ​​is corrected using the probability of failure, i.e.: ; in, ; Obtaining direct temperature deviation ; and Then the temperature judgment is inaccurate; among which, Indicates the radiation deviation threshold. Temperature deviation threshold; When all target temperature sensors are determined to be faulty, the sensorless predictive mode is activated, including: Temperature estimation based on thermodynamic models: Constructing the heat balance equation: ;in, Indicates the internal reference source heat capacity; Indicates the heat dissipation coefficient. Indicates heating power; Indicates ambient temperature. This indicates the predicted temperature of the built-in reference source; Discrete solution: ;in, Indicates the time step.

[0012] Preferably, the construction of the temperature compensation model, which uses ambient temperature and historical data to correct the output thermal image data, includes: Set a sliding time window, acquire all ambient temperatures within the time window, and construct a sequence. ; Set a sliding time window to select multiple built-in reference source temperature data from the built-in reference source temperature sequence to form a historical reference source temperature sequence; Acquire the raw thermal image; Retrieve temperature calibration data of the built-in reference source under normal operating conditions from the database; A temperature compensation model was constructed using a dual-branch CNN-LSTM network architecture. After preprocessing the elements in the ambient temperature sequence, the elements in the reference source temperature sequence, and the original thermal image data, the data are input into the pre-trained temperature compensation model, and the corrected thermal image is output.

[0013] Preferably, the construction of the high-temperature adaptive compensation model and the nonlinear compensation include: Establish mapping function ;in, Represents the original thermal image matrix; Indicates the temperature of the focal plane array; This represents the built-in reference source temperature calibration data vector under normal operating conditions. Indicates model parameters; This represents the compensated thermal image matrix; Represents the compensation function; Generate a temperature drift compensation field, including: Generate global compensation for infrared thermal radiation. ;in Indicates the characteristics of ambient temperature. This represents the temperature characteristics of the focal plane matrix; Indicates feature splicing; Represents the hyperbolic tangent activation function; Represents the global compensation weight matrix; The compensated infrared radiation field is obtained through adaptive fusion. ;in, This indicates the fusion weight. If the ambient temperature is greater than 50℃, then... 1; If the ambient temperature is less than 50℃, then .

[0014] An electronic device includes a processor and a memory and a communication module connected to the processor, the electronic device being used to perform the infrared thermal imaging method based on automatic calibration of sensor parameters according to ambient temperature.

[0015] A computer-readable storage medium storing a computer program that is executed by a processor to implement the infrared thermal imaging method based on automatic calibration of sensor parameters at ambient temperature.

[0016] The beneficial effects of this invention are: This invention establishes a temperature compensation model after a target temperature sensor malfunctions. It adaptively performs temperature compensation using ambient temperature data and historical data (sensor temperature data under normal operating conditions and the original infrared thermal image) to ensure thermal imaging accuracy.

[0017] This invention considers the temperature drift that occurs in temperature sensors and detectors under high-temperature conditions. It integrates a thermodynamic model with a neural network to construct a high-temperature adaptive compensation model to compensate for the temperature drift. Attached Figure Description

[0018] Figure 1 This is a flowchart of the infrared thermal imaging method based on ambient temperature for automatically calibrating sensor parameters according to the present invention. Detailed Implementation

[0019] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious modifications will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0020] It is understood that the term "a" should be understood as "at least one" or "one or more," that is, in one embodiment, the number of an element can be one, while in another embodiment, the number of the element can be multiple, and the term "a" should not be understood as a limitation on the number.

[0021] refer to Figure 1 The technical solution provided by this invention is: an infrared thermal imaging method for automatically calibrating sensor parameters based on ambient temperature, comprising the following steps: Step 1: Obtain temperature data from the built-in reference source to analyze its stability and determine if temperature inaccuracy has occurred. If so, proceed to Step 2; otherwise, proceed to Step 4. The process for determining stability and temperature inaccuracy includes the following steps: Step 1.1: Determine the short-term stability and long-term temperature drift of the built-in reference source to determine if temperature inaccuracy has occurred; specifically: Obtain temperature data from the built-in reference source within a preset time window and construct a temperature sequence from the built-in reference source. ;in, Indicates the length of the time window; Indicates the first time within the time window One built-in reference source temperature; The acquired temperature data from the built-in reference source is compared with the preset temperature threshold. Compare and calculate the temperature deviation within the corresponding time window. , ; Calculating short-term stability includes: Calculate the standard deviation of temperature deviation within the time window. ,if If the short-term stability of the built-in reference source is deemed abnormal, an alarm will be triggered; where, Indicates the standard deviation threshold; Calculate the long-term temperature drift of the built-in reference source, including: Sliding time window, calculation The moving average deviation of temperature deviation over a continuous time window, i.e., the long-term temperature drift. ;in, Indicates the total number of time windows; if If so, the temperature judgment is inaccurate; among them, Temperature deviation threshold.

[0022] Step 1.2: Utilize a time period Long-term temperature drift data were used to construct a long-term drift time series. Step 1.3: Based on the long-term drift time series, predict the future temperature inaccuracy time using a pre-trained trend analysis model; the trend analysis model is an autoregressive integral moving average (ARIMA) model. Specifically: Obtaining the time series of long-term drift , ;in, Indicates the temperature sampling frequency within the time window; right Perform differential processing to obtain the differential time series. ; Fitting the difference time series: ;in, Indicates the order of autoregression. Indicates the order of the moving average. Represents white noise. Indicates the intercept; , Indicates the weighting coefficient; Then, The data is input into a pre-trained trend analysis model; the prediction time is then determined. Long-term temperature drift ; If, | Then determine the time. The temperature was inaccurate.

[0023] Step 2: Calculate the failure probability of the target temperature sensor and determine if the temperature sensor is faulty. If it is, proceed to Step 3; otherwise, proceed to Step 4. The process of calculating the failure probability and determining the temperature sensor failure includes the following steps: The temperature sensor status is predicted and the failure probability is calculated using Kalman filter residual analysis, including: Constructing the state equations: ; ;in, Indicates the target sensor at the current time. The measured temperature value; Indicates the target temperature sensor The temperature value measured at the previous moment; Indicates temperature noise; Indicates observation noise; Using the Kalman filter algorithm, the temperature value at the next moment is predicted based on the sensor's current temperature value. ; Calculate residuals ; The probability of target sensor failure ;in, This represents the standard deviation of the temperature residual under normal operating conditions. if If the target temperature sensor is faulty, then the fault is determined; where, This represents the failure probability threshold.

[0024] Step 3: Fuse sensor failure probabilities to optimize the judgment of inaccuracy of the built-in reference source and predict the temperature of the built-in reference source; then, construct a temperature compensation model and use ambient temperature and historical data to correct the output thermal image data. Specifically, this includes the following steps: Obtain the infrared radiation value of the built-in reference source, and calculate the deviation between the theoretical radiation value and the obtained infrared radiation value of the built-in reference source. ;in, This indicates the radiation value of the built-in reference source. Indicates the theoretical radiation value; ;in, This indicates the temperature of the blackbody. Spectral radiance under; Indicates the emissivity of the built-in reference source; , Indicates the reference source detector gain and offset; The deviation in infrared radiation values ​​is corrected using the probability of failure, i.e.: ; in, ; Obtaining direct temperature deviation ; and Then the temperature judgment is inaccurate; among which, Indicates the radiation deviation threshold. Temperature deviation threshold; When all target temperature sensors are determined to be faulty, the sensorless predictive mode is activated, including: Temperature estimation based on thermodynamic models: Constructing the heat balance equation: ;in, Indicates the internal reference source heat capacity; Indicates the heat dissipation coefficient. Indicates heating power; Indicates ambient temperature. This indicates the predicted temperature of the built-in reference source; Discrete solution: ;in, Indicates the time step.

[0025] A temperature compensation model is constructed to correct the output thermal image data using ambient temperature and historical data, including the following steps: Set a sliding time window, acquire all ambient temperatures within the time window, and construct a sequence: ; Set a sliding time window to select multiple built-in reference source temperature data from the built-in reference source temperature sequence to form a historical reference source temperature sequence; Acquire the raw thermal image; Retrieve temperature calibration data of the built-in reference source under normal operating conditions from the database; A temperature compensation model is constructed using a dual-branch CNN-LSTM network architecture. The LSTM network processes the ambient temperature data and the predicted built-in reference source temperature data. The CNN network corrects the original thermal image by comparing the difference between the built-in reference source temperature data and the ambient temperature data with the difference between the built-in reference source temperature calibration data and the ambient temperature data under normal operating conditions. After preprocessing the elements in the ambient temperature sequence, the elements in the reference source temperature sequence, and the original thermal image data, the data are input into the pre-trained temperature compensation model, and the corrected thermal image is output.

[0026] Step 4: Determine if the ambient temperature exceeds the upper limit of the predicted operating temperature. If so, construct a high-temperature adaptive compensation model and perform nonlinear compensation to ensure temperature measurement accuracy under high-temperature conditions; otherwise, return to Step 1. Specifically: The construction of the high-temperature adaptive compensation model and the nonlinear compensation include: Establish mapping function ;in, Represents the original thermal image matrix; Indicates the temperature of the focal plane array; This represents the built-in reference source temperature calibration data vector under normal operating conditions. Indicates model parameters; This represents the compensated thermal image matrix; Represents the compensation function; Generate a temperature drift compensation field, including: Generate global compensation for infrared thermal radiation. ;in Indicates the characteristics of ambient temperature. This represents the temperature characteristics of the focal plane matrix; Indicates feature splicing; Represents the hyperbolic tangent activation function; Represents the global compensation weight matrix; Among them, the ambient temperature characteristics are obtained from the sequence by an LSTM network. The reference source temperature features are extracted from the reference source temperature sequence by an LSTM network. The compensated infrared radiation field is obtained through adaptive fusion. ;in, This indicates the fusion weight. If the ambient temperature is greater than 50℃, then... 1; If the ambient temperature is less than 50℃, then .

[0027] Because the temperature of the FPA (focal plane matrix) is strongly correlated with the ambient temperature, and this correlation is non-linear, a multi-level compensation strategy is needed to compensate for the FPA temperature in high-temperature environments due to the combined effects of operating conditions and ambient temperature, in order to ensure the accuracy of the FPA temperature under high-temperature conditions.

[0028] The heat conduction model for FPA temperature and ambient temperature is as follows: ; ; Indicates thermal resistance; Indicates the operating current of the FPA; Indicates the equivalent resistance of the FPF; This indicates the temperature rise caused by the detector's self-heating. Indicates the environmental thermal coupling term; ;in, Indicates the thermal coupling coefficient; Represents the thermal time constant. ; This indicates the coupling time. Using the above method, the ambient temperature and its thermal coupling relationship with the FPA (calibrated in the laboratory) are utilized to compensate for the FPA temperature, ensuring the accuracy of the FPA temperature readings.

[0029] Furthermore, the thermodynamic relationship between the FPA temperature, ambient temperature, and built-in reference source temperature is as follows: ;in, This indicates the ambient temperature and the thermally coupled temperature difference. This indicates the set temperature difference of the temperature control system; This indicates the temperature control coefficient; when the temperature sensor malfunctions, ; The supplemented FPA temperature can be used to further compensate for the temperature of the built-in reference source, ensuring the accuracy of temperature measurements.

[0030] The present invention also provides an electronic device, including a processor and a memory and a communication module connected to the processor, the electronic device being used to execute the infrared thermal imaging method based on ambient temperature for automatically calibrating sensor parameters. The present invention also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the infrared thermal imaging method for automatically calibrating sensor parameters based on ambient temperature.

[0031] The processes described above with reference to the flowcharts in the embodiments disclosed in this invention can be implemented as computer software programs. The embodiments disclosed in this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), it performs the functions defined in the methods of this application. It should be noted that the computer-readable medium described above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wire segments, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical fibers, RF, etc., or any suitable combination thereof.

[0032] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0033] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are merely examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functions and structural principles of the present invention have been shown and explained in the embodiments. Without departing from the principles described, the implementation of the present invention may have any changes or modifications.

Claims

1. An infrared thermal imaging method based on automatic calibration of sensor parameters according to ambient temperature, characterized in that, The method includes: Step 1: Obtain temperature data from the built-in reference source to analyze its stability and determine if temperature inaccuracy has occurred; if so, proceed to Step 2; otherwise, proceed to Step 4. Step 2: Calculate the failure probability of the target temperature sensor and determine whether the temperature sensor is faulty. If it is, proceed to Step 3; otherwise, proceed to Step 4. Step 3: Fuse sensor failure probabilities to optimize the judgment of inaccuracy of the built-in reference source and predict the temperature of the built-in reference source; then, build a temperature compensation model and use ambient temperature and historical data to correct the output thermal image data. Step 4: Determine whether the ambient temperature exceeds the upper limit of the predicted operating temperature; if so, construct a high-temperature adaptive compensation model and perform nonlinear compensation to ensure the temperature measurement accuracy under high-temperature conditions; otherwise, return to step 1.

2. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 1, characterized in that, The acquisition of temperature data from the built-in reference source, used to analyze the stability of the built-in reference source and determine whether temperature inaccuracies have occurred, includes: Determine the short-term stability and long-term temperature drift of the built-in reference source to determine if temperature inaccuracy has occurred. Use a time period Long-term temperature drift data were used to construct a long-term drift time series. Based on the long-term drift time series, a pre-trained trend analysis model is used to predict the future temperature inaccuracy time; the trend analysis model is an autoregressive integral moving average (ARIMA) model.

3. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 2, characterized in that, The determination of the short-term stability and long-term temperature drift of the built-in reference source includes: Obtain temperature data from the built-in reference source within a preset time window and construct a temperature sequence from the built-in reference source. ;in, Indicates the length of the time window; Indicates the first time within the time window One built-in reference source temperature; The acquired temperature data from the built-in reference source is compared with the preset temperature threshold. Compare and calculate the temperature deviation within the corresponding time window. , ; Calculating short-term stability includes: Calculate the standard deviation of temperature deviation within the time window. ,if If the short-term stability of the built-in reference source is deemed abnormal, an alarm will be triggered; where, Indicates the standard deviation threshold; Calculate the long-term temperature drift of the built-in reference source, including: Sliding time window, calculation The moving average deviation of temperature deviation over a continuous time window, i.e., the long-term temperature drift. ;in, Indicates the total number of time windows; if If so, the temperature judgment is inaccurate; among them, Temperature deviation threshold.

4. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 3, characterized in that, The method of using a pre-trained trend analysis model to predict future temperature inaccuracies includes: Obtaining the time series of long-term drift , ;in, Indicates the temperature sampling frequency within the time window; right Perform differential processing to obtain the differential time series. ; Fitting the difference time series: ;in, Indicates the order of autoregression. Indicates the order of the moving average. Represents white noise. Indicates the intercept; , Indicates the weighting coefficient; Then, The data is input into a pre-trained trend analysis model; the prediction time is then determined. Long-term temperature drift ; If, | Then determine the time. The temperature was inaccurate.

5. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 4, characterized in that, The calculation of the failure probability of the target temperature sensor and the determination of whether the temperature sensor is faulty include: The temperature sensor status is predicted and the failure probability is calculated using Kalman filter residual analysis, including: Constructing the state equations: ; ;in, Indicates the target sensor at the current time. The measured temperature value; Indicates the target temperature sensor The temperature value measured at the previous moment; Indicates temperature noise; Indicates observation noise; Using the Kalman filter algorithm, the temperature value at the next moment is predicted based on the sensor's current temperature value. ; Calculate residuals ; The probability of target sensor failure ;in, This represents the standard deviation of the temperature residual under normal operating conditions. if If the target temperature sensor is faulty, then the fault is determined; where, This represents the failure probability threshold.

6. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 5, characterized in that, The fusion sensor failure probability optimizes the judgment of inaccuracy of the built-in reference source and predicts the temperature of the built-in reference source, including: Obtain the infrared radiation value of the built-in reference source, and calculate the deviation between the theoretical radiation value and the obtained infrared radiation value of the built-in reference source. ;in, This indicates the radiation value of the built-in reference source. Indicates the theoretical radiation value; ;in, Indicates the temperature of the blackbody Spectral radiance under; Indicates the emissivity of the built-in reference source; , Indicates the reference source detector gain and offset; The deviation in infrared radiation values ​​is corrected using the probability of failure, i.e.: ； in, ; Obtaining direct temperature deviation ; and Then the temperature judgment is inaccurate; among which, Indicates the radiation deviation threshold. Temperature deviation threshold; When all target temperature sensors are determined to be faulty, the sensorless predictive mode is activated, including: Temperature estimation based on thermodynamic models: Constructing the heat balance equation: ;in, Indicates the internal reference source heat capacity; Indicates the heat dissipation coefficient. Indicates heating power; Indicates ambient temperature. This indicates the predicted temperature of the built-in reference source; Discrete solution: ;in, Indicates the time step.

7. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 6, characterized in that, The construction of the temperature compensation model, which uses ambient temperature and historical data to correct the output thermal image data, includes: Set a sliding time window, acquire all ambient temperatures within the time window, and construct a sequence. ; Multiple built-in reference source temperature data are selected from the built-in reference source temperature sequence to form a historical reference source temperature sequence; Obtain the raw thermal image; Retrieve temperature calibration data of the built-in reference source under normal operating conditions from the database; A temperature compensation model was constructed using a dual-branch CNN-LSTM network architecture. After preprocessing the elements in the ambient temperature sequence, the elements in the reference source temperature sequence, and the original thermal image data, the data are input into the pre-trained temperature compensation model, and the corrected thermal image is output.

8. The infrared thermal imaging method based on automatic calibration of sensor parameters according to claim 7, characterized in that, The construction of the high-temperature adaptive compensation model and the nonlinear compensation include: Establish mapping function ;in, Represents the original thermal image matrix; Indicates the temperature of the focal plane array; This represents the built-in reference source temperature calibration data vector under normal operating conditions. Indicates model parameters; This represents the compensated thermal image matrix; Represents the compensation function; Generate a temperature drift compensation field, including: Generate global compensation for infrared thermal radiation. ;in Indicates the characteristics of ambient temperature. This represents the temperature characteristics of the focal plane matrix; Indicates feature splicing; Represents the hyperbolic tangent activation function; Represents the global compensation weight matrix; The compensated infrared radiation field is obtained through adaptive fusion. ;in, This indicates the fusion weight. If the ambient temperature is greater than 50℃, then... 1; If the ambient temperature is less than 50℃, then .

9. An electronic device, comprising a processor and a memory and a communication module connected to the processor, characterized in that, The electronic device is used to perform the infrared thermal imaging method based on ambient temperature for automatically calibrating sensor parameters as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that is executed by a processor to implement the infrared thermal imaging method based on ambient temperature for automatically calibrating sensor parameters as described in any one of claims 1-8.

Images