High-temperature material filter image acquisition and performance testing method for industrial vision

Abstract

The application discloses a high-temperature material filter image collection and performance testing method for industrial vision, and particularly relates to the field of optical imaging technology; the method is characterized in that: original image data of a surface of a measured material is collected, infrared radiation stray light is filtered out, and multi-channel filter image data is obtained; center wavelength drift and bandwidth change of each channel in the multi-channel filter image data are analyzed, and channel spectrum drift data is obtained; brightness stability and edge definition of each channel image are analyzed, and image quality evaluation data is output; a correlation model is established, filter device calibration parameters are output, and filter center wavelengths and bandwidths of each filter device are adjusted in real time, so that calibrated multi-channel filter image data is obtained; a digital speckle correlation method is used to analyze deformation characteristics of the surface of the material, and high-temperature mechanical performance parameters of the material are extracted, which helps to improve the accuracy of material strain identification and mechanical performance extraction.

Description

Technical Field

[0001] This invention relates to the field of optical imaging technology, and more specifically, to a method for high-temperature material filter image acquisition and performance testing for industrial vision. Background Technology

[0002] When testing the mechanical properties of materials in high-temperature environments, it is usually necessary to use composite filter image acquisition technology to obtain clear image data of the material surface.

[0003] When faced with strong infrared radiation emitted from the surface of materials in high-temperature environments, existing composite filtering systems suffer from thermal disturbances in their internal optical components and filtering materials, resulting in optical performance degradation. This leads to uncontrollable dynamic spectral drift in the center wavelength and bandwidth of the filtering band, limiting the accuracy and reliability of mechanical property testing of materials in high-temperature and complex environments. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for high-temperature material filter image acquisition and performance testing for industrial vision to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for high-temperature material filter image acquisition and performance testing for industrial vision includes the following steps: S1: Acquire raw image data of the surface of the material under test in a high-temperature environment, filter out infrared radiation stray light through multi-band composite filtering, and output multi-channel filtered image data; S2: Based on multi-channel filter image data, determine the actual filter center wavelength and actual filter bandwidth of each filter device in the corresponding band, and compare them with the preset filter center wavelength and preset filter bandwidth to obtain channel spectral drift data; S3: Based on multi-channel filtered image data, measure the brightness stability and edge sharpness of each channel filtered image to obtain image quality evaluation data; S4: Establish a mathematical correlation model based on channel spectral drift data and image quality evaluation data, and calculate the target center wavelength compensation amount and target bandwidth compensation amount in combination with the preset image quality target range to obtain the calibration parameters of the filter device; S5: Based on the calibration parameters of the filter devices, compensate and adjust the filtering center wavelength and filtering bandwidth of each filter device in the corresponding band, and collect the calibrated multi-channel filter image data. S6: Based on the calibrated filtered image data, analyze the surface deformation characteristics of the tested material, extract the mechanical property parameters of the tested material, and output the mechanical property evaluation results of the tested material.

[0006] In a preferred embodiment, S1 specifically refers to: A composite filter image acquisition device is used to receive the scattered light from the surface of the material under test in a high-temperature environment and convert it into raw image data; The original image data is subjected to wavelength division filtering by multiple filter devices with different filtering bands to filter out infrared stray light outside each filtering band, and the filtered image data of the corresponding band is obtained. Based on the acquisition timing and spatial correspondence of each filter device, the filter image data of the corresponding band is synchronously aligned and output to form multi-channel filter image data.

[0007] In a preferred embodiment, S2 specifically refers to: Based on multi-channel filter image data, extract the spectral distribution information of each filter device in the corresponding band, and determine the actual filter center wavelength and actual filter bandwidth of the corresponding band; The actual filter center wavelength is compared with the preset filter center wavelength to obtain the center wavelength shift. The actual filter bandwidth is compared with the preset filter bandwidth to obtain the bandwidth change and form channel spectral drift data.

[0008] In a preferred embodiment, S3 specifically refers to: Based on the multi-channel filtered image data, the brightness distribution information and edge grayscale change information of the filtered image data of each filter device in the corresponding band are extracted respectively, and the brightness statistical results and edge sharpness statistical results of the filtered image data of each filter device in the corresponding band are determined. Brightness stability data is determined based on the fluctuation of brightness statistics under continuous acquisition conditions; Edge sharpness data is determined based on the fluctuation of edge sharpness statistics under continuous acquisition conditions; The brightness stability data and edge sharpness data are combined to form image quality evaluation data.

[0009] In a preferred embodiment, S4 specifically refers to: Based on image quality assessment data and channel spectral drift data, the correspondence between center wavelength drift, bandwidth change, brightness stability data and edge sharpness data is extracted. A mathematical correlation model between spectral drift input parameters and image quality target parameters is established based on the correspondence. Based on the mathematical correlation model and combined with the preset image quality target range, the target center wavelength compensation and target bandwidth compensation of each filter device in the corresponding band are calculated to form the calibration parameters of the filter device.

[0010] In a preferred embodiment, S5 specifically refers to: The target center wavelength compensation and target bandwidth compensation for each filter device are extracted based on the calibration parameters of the filter devices. The center wavelength of the filter in the corresponding band of each filter device is compensated and adjusted based on the target center wavelength compensation amount. The filtering bandwidth of each filter device in the corresponding band is compensated and adjusted based on the target bandwidth compensation amount. Based on the compensated and adjusted wavelengths of each filter device, the corresponding bands are synchronously acquired and the calibrated multi-channel filter image data is output.

[0011] In a preferred embodiment, S6 specifically refers to: The displacement change information of the surface of the material under test is extracted based on the calibrated multi-channel filtered image data, and the displacement field data of the surface of the material under test is determined based on the displacement change information. Calculate the strain field data of the surface of the material under test based on the displacement field data of the surface of the material under test; Mechanical property parameters of the tested material are extracted based on strain field data on the surface of the tested material. The mechanical performance evaluation results of the tested material are determined based on its mechanical performance parameters.

[0012] The technical effects and advantages of this invention, a method for high-temperature material filter image acquisition and performance testing for industrial vision: By combining multi-band composite filtering with real-time calibration, high-quality acquisition of material surface images and extraction of mechanical properties are achieved under high-temperature conditions. Multi-band filtering effectively suppresses infrared stray light, significantly improving the signal-to-noise ratio and channel resolution of the original image. Dynamic analysis of center wavelength drift and bandwidth changes allows for real-time quantification of filter performance degradation, ensuring the integrity of spectral data. Combining brightness stability and edge sharpness evaluation comprehensively reflects image quality fluctuations, providing reliable quality indicators for calibration. By establishing a correlation model between spectral drift and image quality, calibration parameters of the filter devices are calculated, ensuring that each band filter maintains optimal filtering performance under high-temperature disturbances. Real-time adjustment of the filter center wavelength and bandwidth yields stable and consistent multi-channel filtered image data, significantly improving image brightness uniformity and edge sharpness. Based on the calibrated filtered image data, digital speckle correlation methods are applied to accurately extract displacement and strain fields, providing high-precision and high-reliability parameter outputs for high-temperature material mechanical property evaluation, greatly improving the accuracy and repeatability of test results. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the high-temperature material filter image acquisition and performance testing method for industrial vision according to the present invention. Detailed Implementation

[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention. Example

[0015] Figure 1 This invention presents a method for high-temperature material filter image acquisition and performance testing for industrial vision, which includes the following steps: S1: Acquire raw image data of the surface of the material under test in a high-temperature environment, filter out infrared radiation stray light through multi-band composite filtering, and output multi-channel filtered image data; S2: Based on multi-channel filter image data, determine the actual filter center wavelength and actual filter bandwidth of each filter device in the corresponding band, and compare them with the preset filter center wavelength and preset filter bandwidth to obtain channel spectral drift data; S3: Based on multi-channel filtered image data, measure the brightness stability and edge sharpness of each channel filtered image to obtain image quality evaluation data; S4: Establish a mathematical correlation model based on channel spectral drift data and image quality evaluation data, and calculate the target center wavelength compensation amount and target bandwidth compensation amount in combination with the preset image quality target range to obtain the calibration parameters of the filter device; S5: Based on the calibration parameters of the filter devices, compensate and adjust the filtering center wavelength and filtering bandwidth of each filter device in the corresponding band, and collect the calibrated multi-channel filter image data. S6: Based on the calibrated filtered image data, analyze the surface deformation characteristics of the tested material, extract the mechanical property parameters of the tested material, and output the mechanical property evaluation results of the tested material.

[0016] S1: Acquires raw image data of the surface of the material under test in a high-temperature environment, filters out infrared stray light through multi-band composite filtering, and outputs multi-channel filtered image data, including: A composite filter image acquisition device is used to receive the scattered light from the surface of the material under test in a high-temperature environment and convert it into raw image data; The composite filtering image acquisition device is a specialized equipment for acquiring images of material surfaces in high-temperature environments. It comprises multiple independently operating filtering devices, an image sensor for recording image information of the material surface, and an optical imaging lens that works in conjunction with the filtering devices and the image sensor. The filtering devices are typically filter lenses or filter arrays, each with a fixed filtering center wavelength and filtering bandwidth. Each filter allows only light within a specific wavelength range to pass through, filtering out all stray light outside that specific wavelength range. For example, a filter with a center wavelength of 550 nm and a bandwidth of 20 nm only allows light within the 540 nm to 560 nm wavelength range to pass through. The image sensor converts the light signal, after stray light has been filtered out by the filtering devices, into digitized image data. For example, it records the optical image of the material surface as raw digitized image data using a charge-coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) device. The optical imaging lens focuses the image information of the material surface onto the image sensor. The raw image data output by the image sensor is unfiltered initial data containing the morphological features and deformation information of the measured material surface under high-temperature conditions. For example, when the surface temperature of the material being tested reaches above 1000 degrees Celsius, the surface typically emits a large amount of infrared stray light. If a composite filter image acquisition device is not used to filter the raw image data, the raw image data recorded by the image sensor will be interfered with by the infrared stray light, resulting in a severe decrease in image contrast or even complete saturation. The temperature range of the high-temperature environment can be from 500 degrees Celsius to 1500 degrees Celsius; the material being tested can be a high-temperature alloy, ceramic material, or silicon carbide composite material. Under the action of external mechanical loads in the high-temperature environment, the surface of the material being tested deforms, thereby generating scattered light with measurable characteristics. The scattered light is recorded by the composite filter image acquisition device to form the raw image data.

[0017] The original image data is subjected to wavelength division filtering by multiple filter devices with different filtering bands to filter out infrared stray light outside each filtering band, and the filtered image data of the corresponding band is obtained. Multiple filter devices with different filtering bands are respectively arranged in the composite filter image acquisition device. Each filter device has a defined filtering center wavelength and filtering bandwidth, and the filtering center wavelengths of any two filter devices are different. For example, when the composite filter image acquisition device is equipped with three filter devices with different filtering bands, the filtering center wavelength of the first filter device is set to 550 nm, the filtering center wavelength of the second filter device is set to 650 nm, and the filtering center wavelength of the third filter device is set to 750 nm. The filtering bandwidth of each of the three filter devices is also clearly defined, for example, all of them are 20 nm, or they are set to 20 nm, 30 nm and 40 nm respectively according to actual test requirements. After the raw image data is filtered sequentially by the three filter devices, each filter device will filter out all infrared radiation stray light outside its own filtering center wavelength, so that the filtered image data of each filter device corresponding to the band contains only the effective light information within the specific band range. The infrared stray light filtered out by each filter mainly originates from the infrared radiation emitted by the surface of the material under test at high temperatures. For example, when the surface of the material under test reaches a temperature of 900 to 1200 degrees Celsius, a large amount of infrared stray light will interfere with the acquisition of image information. The filter devices can effectively filter out the infrared stray light, ensuring that the filtered image data obtained by each filter device is not affected by the infrared stray light. For example, the filtered image data obtained after passing through a 550-nanometer filter has image information within the 550-nanometer ± 10-nanometer wavelength range, reflecting the actual morphology of the material surface within this wavelength range.

[0018] Based on the acquisition timing and spatial correspondence of each filter device, the filter image data of the corresponding band is synchronously aligned and output to form multi-channel filter image data.

[0019] The composite filter image acquisition device is equipped with a data synchronization output device to achieve real-time synchronous output of filter image data from multiple filter devices. This data synchronization output device can be a multi-channel data synchronization transmission device, such as a synchronous data acquisition card or a multi-channel data synchronization transmission device. The filter image data from each filter device is transmitted to the data synchronization output device in real time. After receiving the filter image data output from each filter device, the data synchronization output device performs real-time synchronization alignment processing to ensure that the filter image data output by each filter device is completely consistent in both time and spatial dimensions, avoiding image analysis errors caused by data asynchrony. After processing by the data synchronization output device, the filter image data output by each filter device together constitutes multi-channel filter image data, with each channel corresponding to a specific wavelength of a filter device. Multi-channel filter image data can simultaneously reflect the image information of the measured material surface at multiple wavelengths at the same point in time, effectively avoiding data errors caused by asynchronous image data acquisition. For example, when the surface of the measured material undergoes rapid deformation, the synchronous output of multi-channel filter image data ensures that the image information of each wavelength accurately corresponds to the material surface state at the same deformation moment.

[0020] S2: Based on multi-channel filter image data, determine the actual filter center wavelength and actual filter bandwidth of each filter device in the corresponding band, and compare them with the preset filter center wavelength and preset filter bandwidth to obtain channel spectral drift data, including: Based on multi-channel filter image data, extract the spectral distribution information of each filter device in the corresponding band, and determine the actual filter center wavelength and actual filter bandwidth of the corresponding band; Multi-channel filtered image data is output from a composite filtered image acquisition device and is filtered image data processed by multiple filtered devices with different filtering bands. The filtered image data for each filtered device's corresponding band contains material surface image information after filtering within a specific band. For example, the preset filter center wavelengths of the filtered devices may be different wavelengths such as 550 nm, 650 nm, and 750 nm. The method for determining the actual filter center wavelength is as follows: For the filtered image data for each filtered device's corresponding band, spectral analysis is performed using a spectral measurement device. For example, a spectrometer is used. The spectrometer measures the wavelength of the light signal in the filtered image data and records the light intensity distribution corresponding to each wavelength in the material surface image by scanning wavelength by wavelength, obtaining the spectral intensity distribution curve of the filtered image data output by the filtered devices. The actual filter center wavelength is determined based on the spectral intensity distribution curve. The actual filter center wavelength is the wavelength corresponding to the maximum light intensity value on the spectral intensity distribution curve. For example, if the filter center wavelength of the filter device is calibrated to be 550 nm, and after actual measurement it is found that the light intensity peak of the spectral intensity distribution curve is located at 552 nm, then 552 nm is the actual filter center wavelength.

[0021] The actual filter bandwidth is also determined by measuring the filter image data using a spectrometer. The specific measurement process is as follows: A spectrometer is used to measure the filter image data for each filter device in the corresponding wavelength band, obtaining the spectral intensity distribution curve of the filter image data. On the spectral intensity distribution curve, the peak intensity at the center wavelength of the filter is found. Two wavelength positions are determined at half the peak intensity (i.e., the half-peak), located on the spectral curve on either side of the peak. The difference between the two wavelengths is the actual filter bandwidth. For example, after measuring the filter image data corresponding to a filter device with a center wavelength set to 650 nm, it is found that the peak intensity is located at 649 nm, with a peak intensity of 1000 units. Therefore, the two wavelengths corresponding to an intensity of 500 units are found to be 643 nm and 655 nm, respectively. Thus, the actual filter bandwidth is 655 nm minus 643 nm, which equals 12 nm. The same measurement method is used to determine the actual filter bandwidth of the filter image data for each filter device in the corresponding wavelength band.

[0022] The actual filter center wavelength is compared with the preset filter center wavelength to obtain the center wavelength shift. In a composite filter image acquisition device, each filter has a preset filter center wavelength, such as 550 nm, 650 nm, or 750 nm. In high-temperature environments, the optical performance of the filter may change due to temperature, and the actual filter center wavelength may deviate from the preset wavelength. Therefore, to accurately determine the filtering state of each filter in high-temperature environments, the measured filter center wavelength is compared with the initial preset filter center wavelength. The center wavelength shift is obtained by subtracting the preset wavelength from the actual wavelength. For example, if the preset filter center wavelength is 750 nm, and the actual measurement in a high-temperature environment is 752 nm, then the center wavelength shift for the corresponding band is 752 nm minus 750 nm, which equals 2 nm. By determining and recording the center wavelength shift of all filter devices using this method, the center wavelength shift data of the filter devices is obtained.

[0023] The actual filter bandwidth is compared with the preset filter bandwidth to obtain the bandwidth change and form channel spectral drift data; Optical filters have preset filter bandwidth values, such as 20 nm, 30 nm, or 40 nm. Under high-temperature conditions, changes in the material properties of the filter may cause variations in the actual filter bandwidth. Therefore, to determine the bandwidth variation, the actual measured filter bandwidth is compared with the preset filter bandwidth to determine the amount of bandwidth change. The bandwidth change is the difference between the actual measured bandwidth and the preset filter bandwidth. For example, if the preset filter bandwidth is 30 nm and the actual measured bandwidth is 32 nm, then the bandwidth change is 32 nm minus 30 nm, which equals 2 nm. The bandwidth change is determined for each filter using this method to obtain bandwidth change data.

[0024] The center wavelength shift and bandwidth change reflect the changes in the spectral performance of optical filters under high-temperature conditions. To comprehensively describe the spectral drift of optical filters, the center wavelength shift and bandwidth change are combined to form channel spectral drift data. The combination method is as follows: corresponding combinations are made for each optical filter's corresponding wavelength band. For example, for an optical filter with a center wavelength of 550 nm, if the center wavelength shift is 1 nm and the bandwidth change is 2 nm, the combined channel spectral drift data is expressed as "center wavelength shift of 1 nm and bandwidth change of 2 nm in the 550 nm band". The center wavelength shift and bandwidth change of all optical filters are combined to form complete channel spectral drift data, which is used for calibration and adjustment of optical filters.

[0025] S3: Based on multi-channel filtered image data, measure the brightness stability and edge sharpness of each channel's filtered image to obtain image quality evaluation data, including: Based on the multi-channel filtered image data, the brightness distribution information and edge grayscale change information of the filtered image data of each filter device in the corresponding band are extracted respectively, and the brightness statistical results and edge sharpness statistical results of the filtered image data of each filter device in the corresponding band are determined. Multi-channel filtered image data is generated by a composite filtered image acquisition device processing raw image data separately through multiple filter devices with different filtering bands under high-temperature conditions, and then synchronously outputting the resulting data. Each channel corresponds to a filter device with clearly defined filtering center wavelength and filtering bandwidth. Taking three filtering bands with filtering center wavelengths of 550 nm, 650 nm, and 750 nm as examples, the filtered image data for each filter device band records the actual morphological image information of the material surface within the 550 nm ± 10 nm, 650 nm ± 15 nm, and 750 nm ± 20 nm bands, respectively. The brightness measurement method is as follows: for the filtered image data of each filter device corresponding to a band, the brightness of all pixels in the band filtered image is measured using an image brightness analysis device. The image brightness analysis device is an image analysis instrument used to measure the light intensity distribution of an image, such as an image brightness measurement instrument formed by combining professional image processing software with the equipment. During brightness measurement, the brightness data corresponding to all pixels in the filtered image data is first extracted, and the brightness of each pixel on the material surface image is recorded in digital form. For example, for a filter device with a center wavelength of 550 nanometers, the image size of the corresponding band filter image data is 2048 pixels × 2048 pixels, and the measurement process measures and records the brightness of all 4,194,304 pixels.

[0026] The method for evaluating brightness variation is as follows: By measuring the brightness variation of filtered image data corresponding to the same filter device at multiple consecutive time points, the degree of brightness stability of the filtered image data during continuous acquisition is evaluated. For example, taking the continuous acquisition of 10 frames of filtered image data as an example, with the interval between each frame set to 0.1 seconds, the brightness of each frame of filtered image data is first measured, and the average brightness of each frame is calculated. The average brightness of the 10 frames is recorded sequentially to evaluate the brightness variation of the filtered image data corresponding to the filter device within the continuous time. The calculation method is to subtract the minimum brightness value from the maximum brightness value in the average brightness of the 10 frames, and divide by the average of the average brightness of all 10 frames to obtain the percentage of brightness variation of the filtered image data, which is used to quantify brightness stability. For example, if the maximum average brightness in the 10 frames of images corresponding to the 550 nm filter device is 1000 units, the minimum average brightness is 950 units, and the overall average brightness of the 10 frames is 975 units, then the brightness variation is (1000-950) / 975≈5.13%. The brightness variation amplitude of the filter image data corresponding to the 650 nm and 750 nm filter devices was evaluated using the same method to obtain the brightness stability data of each filter device in the corresponding band.

[0027] Brightness stability data is determined based on the fluctuation of brightness statistics under continuous acquisition conditions; Edge sharpness is measured by quantitatively measuring the edge sharpness of the material surface morphology edge region in the filtered image data corresponding to each filter device in the corresponding wavelength band. The measurement employs image edge sharpness measurement equipment, such as a professional image edge analysis instrument or an edge sharpness analyzer formed by image processing software. Edge sharpness is determined by measuring the magnitude of the gradient of gray-level changes in selected edge regions within the image. For example, in the 550 nm wavelength filtered image data, edge regions of the material surface are selected, and the gray-level distribution curve of the edge region is analyzed to record the magnitude of the gray-level gradient. The larger the gray-level gradient, the higher the edge sharpness. During the measurement process, multiple edge regions of the filtered image data are selected, and multiple edge sharpness values ​​are recorded for each region. Finally, the average edge sharpness of all edge regions is taken as the edge sharpness of the wavelength filtered image data. For example, if the 550 nm filter device corresponds to a band-filtered image data, and 10 edge regions are selected on the material surface, the measured edge sharpness of the 10 edges are 100, 105, 110, 95, 102, 108, 107, 103, 98 and 101, then the edge sharpness of the band-filtered image data is the average of the above edge sharpnesses, which is 102.9.

[0028] Edge sharpness data is determined based on the fluctuation of edge sharpness statistics under continuous acquisition conditions; The evaluation method for edge sharpness variation is as follows: The degree of edge sharpness variation is measured across multiple consecutive filtered image data sets to quantify the stability of edge sharpness during continuous acquisition. For example, after acquiring 10 frames of filtered image data, the edge sharpness of each frame is measured, and the variation in edge sharpness across the 10 frames is calculated. The calculation method is: subtract the minimum edge sharpness value from the maximum value of the 10 frames, and then divide by the average edge sharpness value of the 10 frames. For example, for 550 nm band image data, the maximum edge sharpness value across 10 frames is 110, the minimum is 95, and the average is 102.9. Therefore, the degree of edge sharpness variation is (110-95) / 102.9≈14.57%. The same method is used to evaluate other band image data to determine the edge sharpness data for each filter device in its corresponding band.

[0029] The brightness stability data and edge sharpness data are combined to form image quality evaluation data.

[0030] Brightness stability data and edge sharpness data are image quality indicators for the corresponding wavelength bands of filtered image data for each filter device. To comprehensively evaluate the image quality of the filtered image data for each wavelength band, the brightness stability data and edge sharpness data are combined according to the corresponding wavelength band of the filter device. For example, for the filtered image data for the 550nm filter device, the brightness stability data is 5.13%, and the edge sharpness variation is 14.57%. The combined data is "550nm wavelength band brightness stability data is 5.13%, edge sharpness variation is 14.57%". Other wavelength bands are combined in the same way. The combined data is the image quality evaluation data, used for image quality analysis and filter device adjustment.

[0031] S4: Establish a mathematical correlation model based on channel spectral drift data and image quality evaluation data, and calculate the target center wavelength compensation and target bandwidth compensation based on the preset image quality target range to obtain the calibration parameters of the filter device, including: Based on image quality assessment data and channel spectral drift data, the correspondence between center wavelength drift, bandwidth change, brightness stability data and edge sharpness data is extracted. Using optical filters as the processing object and their corresponding wavelength bands as data indices, a paired sample set is constructed, consisting of spectral drift input parameters and image quality target parameters. Each sample entry in the paired sample set comprises center wavelength drift, bandwidth variation, brightness stability data, and edge sharpness data. The center wavelength drift is obtained by comparing the actual filter center wavelength with the preset filter center wavelength; the bandwidth variation is obtained by comparing the actual filter bandwidth with the preset filter bandwidth; the brightness stability data is calculated from the average brightness fluctuation amplitude of the multi-channel filtered image data at continuous acquisition time points; and the edge sharpness data is calculated from the statistical results of the grayscale change gradient of the multi-channel filtered image data at continuous acquisition time points. The sampling process of the paired sample set is based on continuous acquisition time periods under high-temperature conditions. During the continuous acquisition time period, the multi-channel output of each filter device... The filtered image data forms an image sequence with a consistent time index based on the synchronization alignment results of the data synchronization output device. Each frame in the image sequence corresponds to the measurement results of a spectral measurement device and an image brightness analysis device. The spectral measurement device outputs the actual filter center wavelength and the actual filter bandwidth, while the image brightness analysis device outputs the average brightness and grayscale change gradient statistics. The center wavelength drift, bandwidth change, brightness stability data, and edge sharpness data form the same paired sample entry under the same time index. Paired sample entries are repeatedly recorded under different time indices to form a paired sample set. The size of the paired sample set is determined by the continuous acquisition period length and the number of filter devices. For example, when the continuous acquisition period length is 60 seconds, the acquisition interval is 0.1 seconds, and the number of filter devices is 3, the paired sample set contains 1800 paired sample entries. A mathematical correlation model between spectral drift input parameters and image quality target parameters is established based on the correspondence. The data fitting method selected is either multiple linear regression or polynomial regression. The choice of method is based on the monotonicity and nonlinearity of the changes in luminance stability and edge sharpness data with respect to center wavelength drift and bandwidth variation in the paired sample set. Multiple linear regression is used when the luminance stability and edge sharpness data exhibit a linear trend within the range of center wavelength drift and bandwidth variation; polynomial regression is used when they show a curvature variation within the same range. The multiple linear regression method uses center wavelength drift and bandwidth variation as the input matrix, luminance stability data as the first output vector, and edge sharpness data as the second output vector. The least squares criterion is used to solve for the regression coefficients. The process includes constructing the input matrix, constructing the output vector, calculating the normal equation, and solving the normal equation to obtain the regression coefficients. The regression coefficients include a set of luminance stability regression coefficients and a set of edge sharpness regression coefficients. The polynomial regression method extends the input matrix with quadratic and cross terms. The extended terms include center wavelength drift and bandwidth variation. The extended independent variable input matrix is ​​formed by the squared term of wavelength drift, the squared term of bandwidth change, the product term of center wavelength drift and bandwidth change, the extended term, and the first-order term. The extended independent variable input matrix and the output vector are solved for regression coefficients using the least squares criterion. The effectiveness of the mathematical correlation model is evaluated by residual analysis and hold-out validation. The residual analysis uses the difference between the predicted brightness stability data and the actual brightness stability data, and the difference between the predicted edge sharpness data and the actual edge sharpness data in the paired sample set as the residual sequence. The hold-out validation divides the paired sample set into a training sample set and a validation sample set according to the time index. The training sample set is used to fit the regression coefficients, and the validation sample set is used to calculate the prediction error statistics. The prediction error statistics include mean absolute error and root mean square error. When the mean absolute error and root mean square error meet the preset error threshold, the mathematical correlation model is determined to enter the calibration parameter calculation stage. The preset error threshold is set by using the natural fluctuation range of image quality evaluation data under constant external mechanical load as the reference boundary. The preset error threshold can be set to 0.5 to 1.0 times the natural fluctuation range to avoid overcompensation in the calibration parameter calculation.

[0032] Based on the mathematical correlation model and combined with the preset image quality target range, the target center wavelength compensation and target bandwidth compensation of each filter device in the corresponding band are calculated to form the calibration parameters of the filter device. The image quality target range is defined, which includes the brightness stability target range and the edge sharpness target range. The method for setting the brightness stability target range and the edge sharpness target range adopts a joint determination of baseline calibration and high-temperature constraint. The baseline calibration uses multi-channel filtered image data from the composite filter image acquisition device under low-temperature or normal-temperature environments as the baseline sampling. During baseline sampling, the preset filter center wavelength and preset filter bandwidth remain unchanged, and the external mechanical load remains constant. The baseline sampling yields baseline brightness stability statistics and baseline edge sharpness statistics. The baseline brightness stability statistics include the baseline average brightness and the baseline brightness fluctuation amplitude, and the baseline edge sharpness statistics include the baseline grayscale variation... The gradient mean and the fluctuation range of the baseline grayscale change gradient are used as constraints. High-temperature constraints use the change in surface radiation intensity of the material under high-temperature conditions and the allowable adjustment range of exposure parameters for the composite filter image acquisition device as the constraint boundaries. These constraints limit the target range of image quality to within the achievable range under high-temperature conditions. The target range for brightness stability is set to 1.0 to 2.0 times the baseline brightness fluctuation range, and the target range for edge sharpness is set to 0.8 to 1.0 times the baseline grayscale change gradient mean. Both the brightness stability and edge sharpness target ranges can be recalibrated based on the surface texture scale and speckle pattern contrast of the tested material. For example, when the tested material is a silicon carbide composite material, a higher edge sharpness target can be used. A lower limit for the brightness target is set, while a higher upper limit for the brightness stability target can be used when the tested material is a high-temperature alloy. A calibration parameter solution strategy is constructed, using the target center wavelength compensation and target bandwidth compensation as unknowns, a mathematical correlation model as the constraint equation, and the image quality target range as the constraint condition. The objective function of the calibration parameter solution strategy adopts a deviation minimization form, which uses the squared difference between the predicted brightness stability data and the center value of the brightness stability target range, and the squared difference between the predicted edge sharpness data and the center value of the edge sharpness target range as a weighted sum. The weighting coefficients are set based on the sensitivity of the brightness stability data and edge sharpness data in digital speckle correlation analysis. The contribution ratio and sensitivity contribution ratio are obtained by changing the bandwidth variation under a fixed center wavelength drift and recording the noise level of the displacement field data. The noise level of the displacement field data is characterized by the standard deviation of the displacement field data during the no-load period. The weighting coefficient can be set as the percentage of the increase in the noise level of the displacement field data. The solution method for the calibration parameter solution strategy adopts a combination of grid search and iterative update. The grid search generates a set of candidate points within the feasible interval of the center wavelength drift and bandwidth variation. The feasible interval is obtained by expanding the statistical range of the channel spectral drift data, and the expansion ratio can be set to 1.2 times to 1 times the statistical range.To cover transient drift under high-temperature conditions by 5 times, each candidate point in the candidate point set is input into a mathematical correlation model to obtain predicted brightness stability data and predicted edge sharpness data. When the predicted brightness stability data and predicted edge sharpness data meet the image quality target range constraints, the deviation objective function value is calculated and the candidate point corresponding to the minimum value is recorded. Iterative updates use the candidate point corresponding to the minimum value as the initial value, and continue to search and update the candidate point corresponding to the minimum value by reducing the grid step size. The grid step size reduction method adopts a proportional decreasing strategy, and the decreasing ratio can be set to 0.5 until the grid step size is less than a preset termination threshold. The preset termination threshold can be set to a center wavelength step size and a bandwidth step size in the range of 0.1 nm to 0.5 nm. The target value of center wavelength drift and the target value of bandwidth change corresponding to the candidate point corresponding to the minimum value are used to calculate the target center wavelength compensation and the target bandwidth compensation. The target center wavelength compensation is calculated by making the target center wavelength compensation equal to the target center wavelength drift. The target bandwidth compensation is calculated by multiplying the target bandwidth compensation by the opposite of the target value of the bandwidth change. The target center wavelength compensation and target bandwidth compensation are generated separately for the corresponding wavelength bands of the filter. These two values ​​are combined to form the calibration parameters of the filter and are output. The data structure of the filter calibration parameters uses a triplet list format: filter identifier, target center wavelength compensation, and target bandwidth compensation. The filter identifier is encoded using a combination of the nominal value of the filter center wavelength and the nominal value of the filter bandwidth to avoid filter confusion. For example, a filter with a nominal center wavelength of 550 nm and a nominal bandwidth of 20 nm can be identified as 550-20; a filter with a nominal center wavelength of 650 nm and a nominal bandwidth of 30 nm can be identified as 650-30; and a filter with a nominal center wavelength of 750 nm and a nominal bandwidth of 40 nm can be identified as 750-40.

[0033] S5: Based on the calibration parameters of the filter devices, compensate and adjust the center wavelength and bandwidth of the filter for each corresponding band of the filter device, and acquire the calibrated multi-channel filter image data, including: The target center wavelength compensation and target bandwidth compensation for each filter device are extracted based on the calibration parameters of the filter devices. The target center wavelength compensation amount, based on the corresponding wavelength band of the filter device, is used to adjust the actual center wavelength of the filter device during its current operation. The sign and magnitude of the target center wavelength compensation amount represent the adjustment direction and amplitude, respectively. The adjustment direction and amplitude are determined by the target center wavelength drift value corresponding to the optimal image quality conditions determined by the mathematical correlation model. The calculation method is to take the negative of the target center wavelength drift value. For example, when the target center wavelength drift value of a 550 nm band filter device is determined to be 2.2 nm under high temperature conditions, the target center wavelength compensation amount is -2.2 nm. The negative sign indicates that the actual filtered center wavelength is higher than the preset value, and the actual filtered center wavelength needs to be reduced.

[0034] The center wavelength of the filter in the corresponding band of each filter device is compensated and adjusted based on the target center wavelength compensation amount. The center wavelength of the filter is adjusted using either piezoelectric actuation or thermal expansion. Taking piezoelectric actuation as an example, a piezoelectric actuator is installed outside or inside the optical film or filter lens of the filter device. The voltage drive signal of the piezoelectric actuator is generated by the filter device adjustment controller in the composite filter image acquisition device. The filter device adjustment controller reads the target center wavelength compensation amount from the calibration parameter list and generates a corresponding voltage drive signal based on the sign and amplitude of the target center wavelength compensation amount. The voltage drive signal acts on the piezoelectric actuator, which adjusts the distance between the optical film or filter lens. When the distance between the optical film or filter lens changes, the actual working center wavelength of the filter device changes accordingly. A real-time spectral measurement device measures the center wavelength of the filter device to confirm that the adjusted center wavelength has reached the target value. For example, the piezoelectric actuator uses a piezoelectric ceramic actuator, and the voltage drive signal control range can be set from 0 to 500 volts. The displacement response ratio of the piezoelectric ceramic actuator can be set to 10 nanometers of optical film displacement change per 100 volts.

[0035] The filtering bandwidth of each filter device in the corresponding band is compensated and adjusted based on the target bandwidth compensation amount. The target bandwidth compensation amount is calculated using a mathematical correlation model. The adjustment direction and magnitude are determined by the target value of the bandwidth change, calculated by taking the negative of the target value. For example, when the target value of the bandwidth change for a 650nm filter is set at 1.5nm, the target bandwidth compensation amount is -1.5nm, indicating that the actual filter bandwidth needs to be reduced by 1.5nm to restore the preset filter bandwidth. The actual adjustment of the filter bandwidth is achieved through filter temperature control or film thickness fine-tuning. Taking temperature control as an example, the filter is made of heat-sensitive material, and the filter bandwidth changes with temperature. The implementation method of temperature control is as follows: a precision temperature control element is installed on the outside of the filter; the temperature adjustment signal of the temperature control element is generated by the filter adjustment controller based on the target bandwidth compensation amount; the filter adjustment controller reads the target bandwidth compensation amount from the calibration parameter list, and outputs a temperature control signal according to the sign and magnitude of the target bandwidth compensation amount, driving the temperature control element to raise or lower the temperature of the filter; a real-time spectral measurement device continuously monitors the actual bandwidth of the filter to ensure that the actual bandwidth is consistent with the target bandwidth. For example, the filter device uses a fluoride film system design, which increases the bandwidth by 0.2 nanometers when the temperature increases by 1 degree Celsius. The temperature control element can be set from 20 degrees Celsius to 80 degrees Celsius, and the temperature control accuracy can be set to 0.1 degrees Celsius.

[0036] Based on the compensated and adjusted wavelengths of each filter device, the corresponding bands are synchronously acquired and the calibrated multi-channel filter image data is output.

[0037] The calibrated multi-channel filter image data is the filtered image data after the filter devices have completed real-time compensation for the center wavelength and bandwidth. The acquisition process of the calibrated multi-channel filter image data is as follows: after the filter devices complete real-time compensation and adjustment, the image sensor synchronously acquires the material surface filter image information of multiple filter devices in corresponding wavelength bands through the optical imaging lens; the synchronous output device performs spatiotemporal synchronization alignment processing on the filter image data in real time; the real-time calibrated filter image data ensures that the filter center wavelength and bandwidth of each band are always maintained within the target range. For example, for high-temperature alloy materials under an external load of 1200 degrees Celsius, the real-time calibrated filter center wavelength of the 550 nm filter device is 550 ± 0.2 nm, and the filter bandwidth is 20 ± 0.3 nm; the real-time calibrated filter center wavelength of the 650 nm filter device is 650 ± 0.2 nm, and the filter bandwidth is 30 ± 0.3 nm; and the real-time calibrated filter center wavelength of the 750 nm filter device is 750 ± 0.2 nm, and the filter bandwidth is 40 ± 0.3 nm. The filter device adjustment controller reads the measurement results of the real-time spectral measurement equipment at regular intervals, which can be set between 0.5 seconds and 2 seconds. If the real-time center wavelength or real-time bandwidth exceeds the target range, a new round of real-time calibration and adjustment is initiated to ensure that each filter device maintains a high-precision filtering state throughout the entire high-temperature mechanical test.

[0038] S6: Based on the calibrated filtered image data, analyze the surface deformation characteristics of the tested material, extract the mechanical property parameters of the tested material, and output the mechanical property evaluation results of the tested material, including: The displacement change information of the surface of the material under test is extracted based on the calibrated multi-channel filtered image data, and the displacement field data of the surface of the material under test is determined based on the displacement change information. The calibrated multi-channel filtered image data is acquired by a composite filtered image acquisition device after real-time adjustment and calibration of the filtering center wavelength and filtering bandwidth of each filter device. The multi-channel filtered image data includes material surface image information within multiple filtering bands. For example, it includes material surface filtered image data within multiple bands with center wavelengths of 550 nm, 650 nm, and 750 nm. Digital speckle correlation (DSC) is used for material surface deformation analysis and displacement measurement. Specifically, a speckle pattern with random distribution characteristics is prepared on the surface of the material under test. The speckle pattern is a high-contrast pattern formed by randomly distributed black and white spots, used to accurately record displacement changes in local areas of the material surface. Filtered image data of the material surface before and after external mechanical load is acquired using a composite filtered image acquisition device. Digital speckle correlation analysis software, such as commercial or custom digital speckle image processing software, is used to analyze and process the material surface filtered image data before and after the external mechanical load. The analysis process involves placing the material surface under load... The filtered images before and after the load are divided into several sub-regions, with a defined size for each sub-region, for example, 32 pixels × 32 pixels. Image correlation calculations are performed on the random speckle patterns within each sub-region, calculating the positional offset of the speckle pattern before and after the external load. This positional offset represents the displacement change of the material surface under the external load. By calculating the displacement changes of all sub-regions on the material surface, complete displacement field data of the tested material surface is obtained. The displacement field data is represented by two-dimensional displacement coordinates at various points on the material surface, including both horizontal and vertical displacements. For example, in actual measurements, when the tested material is a silicon carbide composite material, after applying a load at 1000 degrees Celsius, analysis using digital speckle correlation reveals a horizontal displacement of 1.5 micrometers and a vertical displacement of 2.3 micrometers at a certain location on the material surface. The set of displacements from all sub-regions on the material surface constitutes the complete displacement field data.

[0039] Calculate the strain field data of the surface of the material under test based on the displacement field data of the surface of the material under test; Displacement field data represents the displacement coordinates of various points on the material surface. Based on this displacement field data, strain field data of the measured material surface is calculated. Strain field data represents the magnitude of strain in various local regions of the material surface; strain is a quantitative indicator of the degree of deformation of a material under external mechanical loads. The method for calculating strain field data is as follows: The displacement field data of the material surface is mathematically analyzed, for example, by using numerical difference or finite element methods to calculate the spatial derivatives of the displacement at all measurement points on the material surface; based on the strain calculation formula in continuum mechanics, the calculated spatial derivative values ​​of the displacement field data are converted into strain field data; the strain field data includes the transverse strain, longitudinal strain, and shear strain of the material surface; for example, transverse strain represents the degree of deformation in the horizontal direction of the material surface, longitudinal strain represents the degree of deformation in the vertical direction of the material surface, and shear strain represents the degree of local shear deformation that occurs when the material surface undergoes deformation; the displacement field data of the material surface is converted into strain component values ​​for each sub-region; for example, the displacement field data of a certain sub-region is calculated to obtain a transverse strain of 0.0025, a longitudinal strain of 0.0018, and a shear strain of 0.0011. The set of strain component values ​​for all material surface regions is calculated in this way to obtain the complete strain field data of the material surface.

[0040] Mechanical property parameters of the tested material are extracted based on strain field data on the surface of the tested material. The mechanical property parameters of the tested material under high-temperature conditions are indicators that quantify the material's ability to resist deformation and failure under external loads. Strain field data is directly used to calculate and extract these mechanical property parameters. The calculation method is as follows: based on the strain field data, the mechanical property parameters of the tested material, such as the elastic modulus, Poisson's ratio, and shear modulus, are determined. For example, the elastic modulus is calculated based on the ratio between the longitudinal strain and the corresponding applied force in the strain field data; the elastic modulus is the ratio of material stress to longitudinal strain. The Poisson's ratio is calculated based on the ratio between the transverse strain and the longitudinal strain; the Poisson's ratio is the negative ratio of transverse strain to longitudinal strain. The shear modulus is calculated based on the ratio between shear strain and shear stress. These parameters together constitute the mechanical property parameters of the tested material.

[0041] The mechanical performance evaluation results of the tested material are determined based on its mechanical performance parameters.

[0042] The mechanical performance evaluation result is a comprehensive assessment conclusion describing the mechanical properties of the tested material under high-temperature conditions, determined based on mechanical performance parameters. The mechanical performance evaluation result provides quantitative or qualitative assessment data describing the material's resistance to deformation and failure. For example, based on parameters such as elastic modulus, Poisson's ratio, and shear modulus, it is determined that the material has good elastic stability and strength performance at high temperatures. The evaluation result determines that the material is suitable for specific high-temperature engineering applications or that the material's performance is insufficient to meet high-temperature usage requirements. In practical applications, if the tested material is a high-temperature alloy, and measurements show an elastic modulus of 250,000 MPa, a Poisson's ratio of 0.32, and a shear modulus of 80,000 MPa, the mechanical performance evaluation result can be stated as "The high-temperature alloy material exhibits good mechanical properties at 1100 degrees Celsius, with a high elastic modulus and good stability, making it suitable for high-temperature turbine blades and other high-temperature structural components."

[0043] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0044] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0045] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. 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.

[0046] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0047] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0048] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0049] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0050] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 described in the various embodiments of this application. 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.

[0051] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0052] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for high-temperature material filter image acquisition and performance testing for industrial vision, characterized in that, Includes the following steps: S1: Acquire raw image data of the surface of the material under test in a high-temperature environment, filter out infrared radiation stray light through multi-band composite filtering, and output multi-channel filtered image data; S2: Based on multi-channel filter image data, determine the actual filter center wavelength and actual filter bandwidth of each filter device in the corresponding band, and compare them with the preset filter center wavelength and preset filter bandwidth to obtain channel spectral drift data; S3: Based on multi-channel filtered image data, measure the brightness stability and edge sharpness of each channel filtered image to obtain image quality evaluation data; S4: Establish a mathematical correlation model based on channel spectral drift data and image quality evaluation data, and calculate the target center wavelength compensation amount and target bandwidth compensation amount in combination with the preset image quality target range to obtain the calibration parameters of the filter device; S5: Based on the calibration parameters of the filter devices, compensate and adjust the filtering center wavelength and filtering bandwidth of each filter device in the corresponding band, and collect the calibrated multi-channel filter image data. S6: Based on the calibrated filtered image data, analyze the surface deformation characteristics of the tested material, extract the mechanical property parameters of the tested material, and output the mechanical property evaluation results of the tested material.

2. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 1, characterized in that, S1, specifically: A composite filter image acquisition device is used to receive the scattered light from the surface of the material under test in a high-temperature environment and convert it into raw image data; The original image data is subjected to wavelength division filtering by multiple filter devices with different filtering bands to filter out infrared stray light outside each filtering band, and the filtered image data of the corresponding band is obtained. Based on the acquisition timing and spatial correspondence of each filter device, the filter image data of the corresponding band is synchronously aligned and output to form multi-channel filter image data.

3. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 2, characterized in that, S2, specifically: Based on multi-channel filter image data, extract the spectral distribution information of each filter device in the corresponding band, and determine the actual filter center wavelength and actual filter bandwidth of the corresponding band; The actual filter center wavelength is compared with the preset filter center wavelength to obtain the center wavelength shift. The actual filter bandwidth is compared with the preset filter bandwidth to obtain the bandwidth change and form channel spectral drift data.

4. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 3, characterized in that, S3, specifically: Based on the multi-channel filtered image data, the brightness distribution information and edge grayscale change information of the filtered image data of each filter device in the corresponding band are extracted respectively, and the brightness statistical results and edge sharpness statistical results of the filtered image data of each filter device in the corresponding band are determined. Brightness stability data is determined based on the fluctuation of brightness statistics under continuous acquisition conditions; Edge sharpness data is determined based on the fluctuation of edge sharpness statistics under continuous acquisition conditions; The brightness stability data and edge sharpness data are combined to form image quality evaluation data.

5. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 4, characterized in that, S4, specifically: Based on image quality assessment data and channel spectral drift data, the correspondence between center wavelength drift, bandwidth change, brightness stability data and edge sharpness data is extracted. A mathematical correlation model between spectral drift input parameters and image quality target parameters is established based on the correspondence. Based on the mathematical correlation model and combined with the preset image quality target range, the target center wavelength compensation and target bandwidth compensation of each filter device in the corresponding band are calculated to form the calibration parameters of the filter device.

6. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 5, characterized in that, S5, specifically: The target center wavelength compensation and target bandwidth compensation for each filter device are extracted based on the calibration parameters of the filter devices. The center wavelength of the filter in the corresponding band of each filter device is compensated and adjusted based on the target center wavelength compensation amount. The filtering bandwidth of each filter device in the corresponding band is compensated and adjusted based on the target bandwidth compensation amount. Based on the compensated and adjusted wavelengths of each filter device, the corresponding bands are synchronously acquired and the calibrated multi-channel filter image data is output.

7. The method for high-temperature material filter image acquisition and performance testing for industrial vision as described in claim 6, characterized in that, S6, specifically: The displacement change information of the surface of the material under test is extracted based on the calibrated multi-channel filtered image data, and the displacement field data of the surface of the material under test is determined based on the displacement change information. Calculate the strain field data of the surface of the material under test based on the displacement field data of the surface of the material under test; Mechanical property parameters of the tested material are extracted based on strain field data on the surface of the tested material. The mechanical performance evaluation results of the tested material are determined based on its mechanical performance parameters.

Images