A method and device for analyzing temperature measurement error causes of an infrared thermal imager

Abstract

The application discloses a temperature measurement error attribution analysis method and device for an infrared thermal imager. The method comprises the following steps: based on a preset temperature value set of a blackbody radiation source, temperature measurement of the blackbody radiation source is performed by using the infrared thermal imager to obtain a measurement temperature set; image transformation processing is performed on the measurement temperature set to obtain a measurement temperature image set; temperature measurement error attribution analysis processing is performed on the measurement temperature image set to obtain temperature measurement error region information of the infrared thermal imager.

Description

Technical Field

[0001] This invention relates to the fields of industrial data mining and processing, infrared image measurement, and strategy optimization, specifically to a method and apparatus for attribution analysis of temperature measurement errors in infrared thermal imagers. Background Technology

[0002] Infrared thermal imagers, as non-contact temperature measurement devices, have been widely used in many important fields such as industrial equipment fault diagnosis, power system inspection, medical and health monitoring, and security monitoring due to their advantages such as not requiring contact with the target, fast response speed, and ability to measure large areas. The accuracy of their temperature measurement results is directly related to the reliability of subsequent data analysis and the scientific nature of decision-making, and is crucial to ensuring the normal operation of related application scenarios. In actual use, the temperature measurement accuracy of infrared thermal imagers is easily affected by various factors. Among them, the differences in temperature measurement consistency of individual pixels and the inherent nonlinear characteristics of pixel materials are the core reasons for temperature measurement errors. These factors can cause stable systematic errors in some pixel areas of the thermal imager. If these error areas cannot be accurately identified and the source of error cannot be determined, the measurement reliability of the thermal imager will be seriously affected, and even the risk of misjudgment will be triggered.

[0003] Existing technologies for analyzing temperature measurement errors in infrared thermal imagers have significant limitations: most methods only focus on calculating the overall temperature measurement error, making it difficult to accurately locate specific error pixel areas, resulting in a lack of targeted calibration in subsequent steps; some methods rely solely on a single feature for error judgment, failing to comprehensively consider the grayscale distribution patterns and shape and structural characteristics of the area, leading to insufficient accuracy in identifying error areas; furthermore, existing technologies often fail to fully integrate the error distribution patterns under different temperature conditions, making it difficult to distinguish between random errors and systematic errors caused by pixel characteristics, resulting in incomplete and unreliable error attribution analysis. This fails to provide effective data support for the accurate calibration of infrared thermal imagers, thus restricting the application of infrared thermal imagers in high-precision temperature measurement scenarios. Summary of the Invention

[0004] This invention primarily addresses the problem that error attribution analysis in the calibration field of infrared thermal imagers is not comprehensive or reliable enough, failing to provide effective data support for the accurate calibration of infrared thermal imagers and thus restricting the application of infrared thermal imagers in high-precision temperature measurement scenarios. This invention discloses a method and apparatus for attribution analysis of temperature measurement errors in infrared thermal imagers.

[0005] In a first aspect, this invention discloses a method for attributing temperature measurement errors in an infrared thermal imager, comprising:

[0006] S1, based on a preset set of temperature values ​​of a blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain a set of measured temperatures; the set of temperature values ​​includes several temperatures of the blackbody radiation source.

[0007] S2, perform image transformation processing on the measured temperature set to obtain a measured temperature image set;

[0008] S3, Perform temperature measurement error attribution analysis on the set of measured temperature images to obtain temperature measurement error area information of the infrared thermal imager;

[0009] The temperature measurement error region information is used to characterize the main pixel region information that generates temperature measurement errors in infrared thermal imagers, which are caused by inconsistencies in pixel temperature measurement or nonlinearity errors in the range of pixel materials.

[0010] The set of temperature values ​​for a preset blackbody radiation source is obtained by measuring the temperature of the blackbody radiation source using an infrared thermal imager, and includes:

[0011] S11, Based on a preset set of temperature values ​​for blackbody radiation sources, set the temperature of the blackbody radiation source so that the blackbody radiation source fills the field of view of the infrared thermal imager.

[0012] S12, at each temperature of the blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain the temperature values ​​of all pixels of the infrared thermal imager; using the temperature values ​​of all pixels, a subset of measured temperatures corresponding to the temperature of the blackbody radiation source is constructed.

[0013] S13, using the temperature values ​​of all pixels of the infrared thermal imager obtained at all temperatures of the blackbody radiation source, a set of measurement temperatures is constructed; the set of measurement temperatures includes a subset of measurement temperatures corresponding to the temperature of each blackbody radiation source.

[0014] The step of performing image transformation processing on the measured temperature set to obtain a measured temperature image set includes:

[0015] S21, For each subset of the measured temperatures in the set of measured temperatures, subtract the temperature of the corresponding blackbody radiation source to obtain the corresponding subset of the difference temperatures;

[0016] S22, for each subset of temperature differences, it is represented as a corresponding measured temperature image; the gray value of each pixel in the measured temperature image is the temperature difference of the corresponding pixel in the subset of temperature differences; the coordinates of each pixel in the measured temperature image are determined by the coordinates of the corresponding pixel in the subset of temperature differences.

[0017] S23. Using all the measured temperature images, construct a set of measured temperature images.

[0018] The step of performing temperature measurement error attribution analysis on the set of measured temperature images to obtain temperature measurement error region information of the infrared thermal imager includes:

[0019] S31, each temperature measurement image in the set of temperature measurement images is evenly divided into several sub-images;

[0020] S32, for each sub-image in the measured temperature image, perform matching processing with the sub-images of all other measured temperature images in the set of measured temperature images to obtain the corresponding set of matched sub-images; the set of matched sub-images includes several matched sub-images;

[0021] S33, statistically obtain the set of matching sub-images with the largest number of matching sub-images; determine the positional region information of the sub-images corresponding to the statistically obtained set of matching sub-images in the measured temperature image, which is the temperature measurement error region information of the infrared thermal imager.

[0022] The process of matching each sub-image in the measured temperature image with sub-images of all other measured temperature images in the set of measured temperature images to obtain a corresponding set of matched sub-images includes:

[0023] S321, For each sub-image in each measured temperature image, feature information is extracted to obtain the feature information set corresponding to the sub-image;

[0024] S322, For the feature information set of all sub-images in each measured temperature image, perform filtering and detection processing respectively to obtain the filtered sub-image set;

[0025] S323, For each sub-image in the filtered sub-image set of the measured temperature image, perform matching processing with the sub-images in the filtered sub-image sets of all other measured temperature images in the measured temperature image set to obtain the corresponding matching sub-image set.

[0026] The step involves extracting feature information from each sub-image within each measured temperature image to obtain a set of feature information corresponding to the sub-image, including:

[0027] S3211, For each sub-image in each measured temperature image, extract texture features to obtain the texture feature information corresponding to the sub-image; the texture feature information includes the angular second distance, contrast value and entropy value of the sub-image;

[0028] S3212, For each sub-image in each measured temperature image, shape and structural features are extracted to obtain the structural feature information corresponding to the sub-image; the structural feature information includes the gray values ​​of the corner points and the gray values ​​of the edge points of the sub-image;

[0029] S3213, using all the texture and structural features of a sub-image, construct the feature information set corresponding to the sub-image.

[0030] The feature information set of all sub-images in each measured temperature image is filtered and detected to obtain a filtered set of sub-images, including:

[0031] S3221, calculate the difference between the feature information set of each sub-image in each measured temperature image and the preset standard temperature measurement image feature information set to obtain the difference value corresponding to the sub-image;

[0032] S3222, using all the sub-images of each measured temperature image whose difference value is greater than a preset discrimination threshold, a set of filtered sub-images of the measured temperature image is constructed.

[0033] The expression for calculating the degree of difference is:

[0034] ,

[0035] ,

[0036] Where A represents intermediate computational cost. and These are preset calculation constants. C Here, N represents the difference value, and N is the total number of elements contained in the feature information set of the sub-image. and Let i be the i-th element of the feature information set of the sub-image and the feature information set of the standard temperature measurement image, respectively. The weight value of the i-th element is preset; the first to fifth elements of the feature information set are the second angular distance, contrast value, entropy value, corner gray value and edge gray value, respectively.

[0037] A second aspect of this invention discloses a device for attributing temperature measurement errors in an infrared thermal imager, the device comprising:

[0038] Memory containing executable program code;

[0039] A processor coupled to the memory;

[0040] The processor calls the executable program code stored in the memory to execute the infrared thermal imager's temperature measurement error attribution analysis method.

[0041] In a third aspect of this invention, a computer-readable storage medium is disclosed, wherein the computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the temperature measurement error attribution analysis method of the infrared thermal imager.

[0042] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the temperature measurement error attribution analysis method of the infrared thermal imager.

[0043] The beneficial effects of this invention are as follows:

[0044] This invention, by pre-setting a blackbody radiation source temperature set containing multiple temperature values ​​and filling the infrared thermal imager's field of view with the blackbody radiation source during each measurement, can comprehensively collect measurement data of all pixels of the thermal imager under different temperature conditions. This avoids the one-sidedness of local measurement data and provides complete and comprehensive basic data for subsequent error analysis, ensuring that error analysis covers the entire temperature measurement area of ​​the thermal imager.

[0045] This invention calculates the difference between the measured data at each temperature and the standard temperature of the corresponding blackbody radiation source, and converts the difference result into a measured temperature image. This allows the temperature measurement error of each pixel to be presented in an intuitive image form, clearly reflecting the spatial distribution characteristics of the error, and providing a convenient and visual basis for the subsequent location and analysis of the error area.

[0046] In the feature extraction process, this invention comprehensively acquires the texture features and shape structure features of the sub-image. The texture features can reflect the uniformity, contrast and complexity of the gray-level distribution of the region, while the shape structure features can capture the key inflection points and contour information of the region. The combination of multi-dimensional features comprehensively describes the essential characteristics of the sub-image, greatly improves the accuracy of error region screening and matching, and effectively avoids misjudgment caused by a single feature.

[0047] This invention calculates the difference between the feature information of sub-images and preset standard feature information, and filters out sub-images with significant differences based on preset discrimination thresholds. This can quickly eliminate areas with normal temperature measurement, accurately locate potential error areas, narrow the scope of subsequent matching analysis, and improve the efficiency of overall error attribution analysis. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0049] To better understand the content of this invention, an embodiment is provided here.

[0050] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0051] In a first aspect, this invention discloses a method for attributing temperature measurement errors in an infrared thermal imager, comprising:

[0052] S1, based on a preset set of temperature values ​​of a blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain a set of measured temperatures; the set of temperature values ​​includes several temperatures of the blackbody radiation source.

[0053] S2, perform image transformation processing on the measured temperature set to obtain a measured temperature image set;

[0054] S3, Perform temperature measurement error attribution analysis on the set of measured temperature images to obtain temperature measurement error area information of the infrared thermal imager;

[0055] The temperature measurement error region information is used to characterize the main pixel region information that generates temperature measurement errors in infrared thermal imagers, which are caused by inconsistencies in pixel temperature measurement or nonlinearity errors in the range of pixel materials.

[0056] The set of temperature values ​​for a pre-defined blackbody radiation source is used to measure the temperature of the blackbody radiation source using an infrared thermal imager, resulting in a set of measured temperatures.

[0057] S11, Based on a preset set of temperature values ​​for blackbody radiation sources, set the temperature of the blackbody radiation source so that the blackbody radiation source fills the field of view of the infrared thermal imager.

[0058] S12, at each temperature of the blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain the temperature values ​​of all pixels of the infrared thermal imager; using the temperature values ​​of all pixels, a subset of measured temperatures corresponding to the temperature of the blackbody radiation source is constructed.

[0059] S13, using the temperature values ​​of all pixels of the infrared thermal imager obtained at all temperatures of the blackbody radiation source, a set of measurement temperatures is constructed; the set of measurement temperatures includes a subset of measurement temperatures corresponding to the temperature of each blackbody radiation source.

[0060] The step of performing image transformation processing on the measured temperature set to obtain a measured temperature image set includes:

[0061] S21, For each subset of the measured temperatures in the set of measured temperatures, subtract the temperature of the corresponding blackbody radiation source to obtain the corresponding subset of the difference temperatures;

[0062] S22, for each subset of temperature differences, it is represented as a corresponding measured temperature image; the gray value of each pixel in the measured temperature image is the temperature difference of the corresponding pixel in the subset of temperature differences; the coordinates of each pixel in the measured temperature image are determined by the coordinates of the corresponding pixel in the subset of temperature differences.

[0063] S23. Using all the measured temperature images, construct a set of measured temperature images.

[0064] Subtracting each of the measured temperature subsets from the temperature of the corresponding blackbody radiation source involves subtracting each element of each measured temperature subset from the temperature of the corresponding blackbody radiation source to obtain the corresponding difference temperature. Using all the difference temperatures, a difference temperature subset is constructed.

[0065] The step of performing temperature measurement error attribution analysis on the set of measured temperature images to obtain temperature measurement error region information of the infrared thermal imager includes:

[0066] S31, each temperature measurement image in the set of temperature measurement images is evenly divided into several sub-images;

[0067] S32, for each sub-image in the measured temperature image, perform matching processing with the sub-images of all other measured temperature images in the set of measured temperature images to obtain the corresponding set of matched sub-images; the set of matched sub-images includes several matched sub-images;

[0068] S33, statistically obtain the set of matching sub-images with the largest number of matching sub-images; determine the positional region information of the sub-images corresponding to the statistically obtained set of matching sub-images in the measured temperature image, which is the temperature measurement error region information of the infrared thermal imager.

[0069] The location region information refers to the region information of the position coordinates of the pixels in the sub-image, which corresponds to the region information of the position coordinates of the pixels in the infrared thermal imager.

[0070] The process of matching each sub-image in the measured temperature image with sub-images of all other measured temperature images in the set of measured temperature images to obtain a corresponding set of matched sub-images includes:

[0071] S321, For each sub-image in each measured temperature image, feature information is extracted to obtain the feature information set corresponding to the sub-image;

[0072] S322, For the feature information set of all sub-images in each measured temperature image, perform filtering and detection processing respectively to obtain the filtered sub-image set;

[0073] S323, For each sub-image in the filtered sub-image set of the measured temperature image, perform matching processing with the sub-images in the filtered sub-image sets of all other measured temperature images in the measured temperature image set to obtain the corresponding matching sub-image set.

[0074] The step involves extracting feature information from each sub-image within each measured temperature image to obtain a set of feature information corresponding to the sub-image, including:

[0075] S3211, For each sub-image in each measured temperature image, extract texture features to obtain corresponding texture feature information; the texture feature information includes the angular second moment (energy), contrast value, and entropy value of the sub-image.

[0076] S3212, For each sub-image in each measured temperature image, shape and structural features are extracted to obtain corresponding structural feature information; the structural feature information includes the gray values ​​of corner points and edge points of the sub-image;

[0077] S3213: Using all the texture and structural features of a sub-image, construct the corresponding feature information set.

[0078] The feature information set of all sub-images in each measured temperature image is filtered and detected to obtain a filtered set of sub-images, including:

[0079] For each sub-image feature information set in each measured temperature image, the difference degree is calculated with the preset standard temperature measurement image feature information set to obtain the difference value corresponding to the sub-image.

[0080] By using all sub-images of a temperature measurement image whose difference values ​​are greater than a preset discrimination threshold, a filtered set of sub-images of the temperature measurement image is constructed.

[0081] The preset set of standard temperature measurement image feature information includes standard angular second distance, standard contrast value, standard entropy value, standard corner gray value and standard edge gray value;

[0082] The expression for calculating the degree of difference is:

[0083] ,

[0084] ,

[0085] Where A represents intermediate computational cost. and These are preset calculation constants. C Here, N represents the difference value, and N is the total number of elements contained in the feature information set of the sub-image. and Let i be the i-th element of the feature information set of the sub-image and the feature information set of the standard temperature measurement image, respectively. The weight value of the i-th element is preset; the first to fifth elements of the feature information set are the second angular distance, contrast value, entropy value, corner gray value and edge gray value, respectively.

[0086] The expression for calculating the difference, through a combination of sine and exponential functions, constructs a nonlinear mapping relationship for the difference value, possessing multiple advantages: First, the sine function part can normalize and constrain the ratio of intermediate calculation quantities to standard values, ensuring that the difference value is always within a reasonable range and avoiding numerical overflow or distortion caused by extreme deviations; Second, the exponential function part can dynamically adjust the difference response sensitivity, compressing small deviations close to the standard state and amplifying significant deviations, thereby accurately distinguishing between normal fluctuations and abnormal deviations; Third, the overall function shape balances the accuracy and robustness of difference judgment, effectively identifying characteristic abnormal sub-images caused by pixel inconsistencies or range nonlinearity, while also resisting interference from random noise, providing a scientific and rigorous quantitative basis for screening out truly error-related sub-images.

[0087] The intermediate computational expression, by weighting and square-rooting the differences between the features of a single sub-image and the standard features, comprehensively reflects the overall deviation of the sub-image from the standard state across all key feature dimensions. On one hand, weighting highlights the differences in importance of different features in representing temperature measurement errors, avoiding the interference of minor deviations in secondary features with the overall judgment. On the other hand, the combination of square and square-root operations amplifies the influence of significantly deviating features while smoothing out small random errors, ensuring that the intermediate computational results accurately capture the core differences between the sub-image features and the standard state, providing reliable basic data support for subsequent difference value calculations.

[0088] The texture feature extraction can be achieved by calculating the second angular distance (energy), contrast value, and entropy value.

[0089] The extraction of shape and structural features can be achieved by using methods for extracting corner points and edge points.

[0090] The sub-images in the filtered sub-image set for each measured temperature image are matched with the sub-images in the filtered sub-image sets of all other measured temperature images in the measured temperature image set to obtain corresponding matched sub-image sets, including:

[0091] For each sub-image in the filtered sub-image set of each measured temperature image, initialize the corresponding matching sub-image set;

[0092] For the sub-image, a matching process is performed with each sub-image in the filtered sub-image set of all other temperature measurement images in the temperature measurement image set to obtain the matching result of each sub-image in the filtered sub-image set of all other temperature measurement images;

[0093] All sub-images that match are added to the initial set of matched sub-images for the sub-image, thus obtaining the set of matched sub-images corresponding to the sub-image.

[0094] The matching process includes:

[0095] For the sub-image, and the sub-images of the set of sub-images filtered from other temperature measurement images, pixel grayscale value statistical processing is performed to obtain the corresponding cumulative probability density function;

[0096] The cumulative probability density functions of the two sub-images are used to calculate the functional difference, and the corresponding matching values ​​are obtained.

[0097] Determine whether the matching value is greater than a preset matching threshold to obtain a first discrimination result; if the first discrimination result is greater than, confirm that the matching result of the sub-image in the set of sub-images after filtering other temperature measurement images is a match; if the first discrimination result is not greater than, confirm that the matching result of the sub-image in the set of sub-images after filtering other temperature measurement images is a mismatch.

[0098] The pixel grayscale value statistical processing includes:

[0099] For each sub-image, its gray-level histogram is obtained; when the pixel gray-level value of the sub-image is d, the corresponding gray-level histogram value is... The calculation formula is:

[0100] ,

[0101] In the formula, P is the total number of pixels in the sub-image. The total number of pixels with a gray value of d in the sub-image; a gray-level histogram is constructed using the gray-level histogram values ​​of all pixel gray values;

[0102] Based on the gray-level histogram, the cumulative probability density function is calculated; when the pixel gray-level value is... d When, the corresponding cumulative probability density function for:

[0103] ,

[0104] in, This represents the grayscale histogram value when the pixel grayscale value is l.

[0105] The cumulative probability density function, by summing the gray-level histogram, transforms the probability distribution of gray-level values ​​into a cumulative distribution feature, further deepening the characterization of the temperature difference distribution. On one hand, the cumulative operation can highlight the overall trend of the gray-level distribution of the sub-image. For example, rapid accumulation indicates a high proportion of pixels with low temperature differences, while slow accumulation indicates a more concentrated distribution of pixels with high temperature differences, thus intuitively reflecting the error distribution characteristics of the sub-image. On the other hand, the cumulative probability density function transforms the discrete gray-level distribution into a continuous distribution curve, which not only facilitates subsequent function-level difference comparisons but also effectively smooths local fluctuations in the gray-level histogram, reduces the interference of isolated noise pixels on the matching results, and improves the stability and reliability of the matching process.

[0106] The expression for calculating the difference of the functions is:

[0107]

[0108] in, pq For the matching value, and Let be the cumulative probability density function of the two sub-images, and q be the pixel gray value. It is a second-order Weber function. This represents the maximum value of the pixel's grayscale value.

[0109] The expression for calculating the functional difference achieves precise quantification of the similarity of grayscale distributions between the two sub-images by integrating the cumulative probability density functions of the two sub-images and adjusting the weights of the second-order Weber function. Firstly, the integration operation comprehensively covers all grayscale value ranges, avoiding misjudgments of the overall matching result due to local grayscale differences, and ensuring that the matching value reflects the consistency of the overall temperature distribution pattern of the sub-images. Secondly, the introduction of the second-order Weber function allows for dynamic adjustment of weights based on different grayscale values, focusing on grayscale ranges sensitive to human vision and temperature measurement errors, making the matching result more consistent with actual error characterization needs. Thirdly, by calculating and integrating the absolute difference between the two functions, the difference in distribution patterns can be transformed into an intuitive numerical indicator. The larger this indicator, the more similar the error distribution patterns of the two sub-images, providing an objective and comparable quantitative standard for judging whether sub-images under different temperature conditions belong to the same error region, ultimately helping to accurately locate the main temperature measurement error regions of the infrared thermal imager.

[0110] This invention achieves accurate attribution of systematic errors by performing cross-image matching between the selected sub-images and all other selected sub-images corresponding to different temperatures, and by statistically analyzing the set of sub-images with the most matches. This enables the precise identification of pixel regions with stable errors under different temperature conditions, thereby effectively distinguishing between random errors and regular errors caused by pixel characteristics, and ensuring the reliability of error attribution results.

[0111] This invention clarifies that the formation of temperature measurement error areas originates from inconsistencies in pixel temperature measurement or nonlinearity errors in the range of pixel materials. The output error area location information can provide precise guidance for targeted calibration of infrared thermal imagers, helping technicians to quickly locate the pixel areas that need correction. Through directional calibration, the overall temperature measurement accuracy of infrared thermal imagers is significantly improved, expanding their application boundaries in high-precision temperature measurement scenarios and enhancing the market competitiveness of the products.

[0112] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.

[0113] In all embodiments of the present invention, the values ​​of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.

[0114] A second aspect of this invention discloses a device for attributing temperature measurement errors in an infrared thermal imager, the device comprising:

[0115] Memory containing executable program code;

[0116] A processor coupled to the memory;

[0117] The processor calls the executable program code stored in the memory to execute the infrared thermal imager's temperature measurement error attribution analysis method.

[0118] In a third aspect of this invention, a computer-readable storage medium is disclosed, wherein the computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the temperature measurement error attribution analysis method of the infrared thermal imager.

[0119] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the temperature measurement error attribution analysis method of the infrared thermal imager.

[0120] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for attributing temperature measurement errors in an infrared thermal imager, characterized in that, include: S1, based on a preset set of temperature values ​​of a blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain a set of measured temperatures; the set of temperature values ​​includes several temperatures of the blackbody radiation source. S2, perform image transformation processing on the measured temperature set to obtain a set of measured temperature images, including: S21, For each subset of the measured temperatures in the set of measured temperatures, subtract the temperature of the corresponding blackbody radiation source to obtain the corresponding subset of the difference temperatures; S22, for each subset of temperature differences, it is represented as a corresponding measured temperature image; the gray value of each pixel in the measured temperature image is the temperature difference of the corresponding pixel in the subset of temperature differences; the coordinates of each pixel in the measured temperature image are determined by the coordinates of the corresponding pixel in the subset of temperature differences. S23, using all the measured temperature images, construct a set of measured temperature images; S3, perform temperature measurement error attribution analysis on the set of measured temperature images to obtain temperature measurement error region information of the infrared thermal imager, including: S31, each temperature measurement image in the set of temperature measurement images is evenly divided into several sub-images; S32, for each sub-image in the measured temperature image, perform matching processing with the sub-images of all other measured temperature images in the set of measured temperature images to obtain the corresponding set of matched sub-images; the set of matched sub-images includes several matched sub-images; S33, statistically obtain the set of matching sub-images with the largest number of matching sub-images; determine the positional region information of the sub-images corresponding to the statistically obtained set of matching sub-images in the measured temperature image, which is the temperature measurement error region information of the infrared thermal imager.

2. The method for attributing temperature measurement errors in an infrared thermal imager as described in claim 1, characterized in that, The set of temperature values ​​for a preset blackbody radiation source is obtained by measuring the temperature of the blackbody radiation source using an infrared thermal imager, and includes: S11, Based on a preset set of temperature values ​​for blackbody radiation sources, set the temperature of the blackbody radiation source so that the blackbody radiation source fills the field of view of the infrared thermal imager. S12, at each temperature of the blackbody radiation source, the temperature of the blackbody radiation source is measured using an infrared thermal imager to obtain the temperature values ​​of all pixels of the infrared thermal imager; using the temperature values ​​of all pixels, a subset of measured temperatures corresponding to the temperature of the blackbody radiation source is constructed. S13, using the temperature values ​​of all pixels of the infrared thermal imager obtained at all temperatures of the blackbody radiation source, a set of measurement temperatures is constructed; the set of measurement temperatures includes a subset of measurement temperatures corresponding to the temperature of each blackbody radiation source.

3. The method for attributing temperature measurement errors in an infrared thermal imager as described in claim 1, characterized in that, The process of matching each sub-image in the measured temperature image with sub-images of all other measured temperature images in the set of measured temperature images to obtain a corresponding set of matched sub-images includes: S321, For each sub-image in each measured temperature image, feature information is extracted to obtain the feature information set corresponding to the sub-image; S322, For the feature information set of all sub-images in each measured temperature image, perform filtering and detection processing respectively to obtain the filtered sub-image set; S323, For each sub-image in the filtered sub-image set of the measured temperature image, perform matching processing with the sub-images in the filtered sub-image sets of all other measured temperature images in the measured temperature image set to obtain the corresponding matching sub-image set.

4. The method for attributing temperature measurement errors in an infrared thermal imager as described in claim 3, characterized in that, The step involves extracting feature information from each sub-image within each measured temperature image to obtain a set of feature information corresponding to the sub-image, including: S3211, For each sub-image in each measured temperature image, extract texture features to obtain the texture feature information corresponding to the sub-image; the texture feature information includes the angular second distance, contrast value and entropy value of the sub-image; S3212, For each sub-image in each measured temperature image, shape and structural features are extracted to obtain the structural feature information corresponding to the sub-image; the structural feature information includes the gray values ​​of the corner points and the gray values ​​of the edge points of the sub-image; S3213, using all the texture and structural features of a sub-image, construct the feature information set corresponding to the sub-image.

5. The method for attributing temperature measurement errors in an infrared thermal imager as described in claim 3, characterized in that, The feature information set of all sub-images in each measured temperature image is filtered and detected to obtain a filtered set of sub-images, including: S3221, calculate the difference between the feature information set of each sub-image in each measured temperature image and the preset standard temperature measurement image feature information set to obtain the difference value corresponding to the sub-image; S3222, using all the sub-images of each measured temperature image whose difference value is greater than a preset discrimination threshold, a set of filtered sub-images of the measured temperature image is constructed; The expression for calculating the degree of difference is: ， ， Where A represents intermediate computational cost. and These are preset calculation constants. C Here, N represents the difference value, and N is the total number of elements contained in the feature information set of the sub-image. and Let i be the i-th element of the feature information set of the sub-image and the feature information set of the standard temperature measurement image, respectively. The weight value of the i-th element is preset; the first to fifth elements of the feature information set are the second angular distance, contrast value, entropy value, corner gray value and edge gray value, respectively.

6. A device for attributing temperature measurement errors in an infrared thermal imager, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the temperature measurement error attribution analysis method of the infrared thermal imager as described in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the temperature measurement error attribution analysis method of the infrared thermal imager as described in any one of claims 1 to 5.

8. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the temperature measurement error attribution analysis method of the infrared thermal imager as described in any one of claims 1 to 5.

Images