Thermal imaging data conversion method, modulation signal reconstruction method and computer device
By establishing a conversion model between modulated thermal imaging and flash thermal imaging data, the problems of insufficient sensitivity of flash thermal imaging to deep defects and blind frequency phenomenon of modulated thermal imaging are solved, achieving efficient depth detection and resolution optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CAPITAL NORMAL UNIVERSITY
- Filing Date
- 2025-12-12
- Publication Date
- 2026-05-12
AI Technical Summary
In existing technologies, flash thermal imaging has limited sensitivity to deep defects, modulated thermal imaging is prone to blind frequency phenomenon at specific frequencies, and traditional modulated thermal imaging requires multiple experiments, which increases the complexity and time consumption of detection.
By establishing a conversion model between modulated thermal imaging and flash thermal imaging data, the flash temperature curve is reconstructed using the modulated temperature data, and the steady-state thermal wave response of any modulation frequency is reconstructed using the flash temperature, thus realizing bidirectional data conversion and signal reconstruction.
Based on single-modulation thermal imaging experiments, the need for multiple experiments is reduced, the ability to reveal deep defects is improved, missed detections caused by blind frequency are avoided, detection efficiency is optimized, and the balance between detection depth and resolution is taken into account.
Smart Images
Figure CN122016930A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of active infrared thermal imaging nondestructive testing technology, and particularly relates to a method for converting modulated thermal imaging data and flash thermal imaging data, a signal reconstruction method, and a computer device. It is suitable for bidirectional conversion of modulated thermal imaging data and flash thermal imaging data and modulation response reconstruction, and is used for quantitative detection and imaging of internal defects in materials. Background Technology
[0002] Active infrared thermography is a non-destructive testing technique that utilizes external thermal excitation to record the temperature field changes of the surface of the object under test over time, and then infers the internal defects, damage, and other anisotropic structures of the material through inversion. This method features non-contact operation, rapid imaging, and the ability to cover large areas, and is widely used in aerospace, rail transportation, energy equipment, composite material manufacturing, and structural health monitoring. Commonly used active excitation methods mainly include flash thermography and modulated thermography, which differ in their excitation methods, signal characteristics, and applicable depth.
[0003] Flash thermal imaging applies energy to a material surface instantaneously via a short-pulse heat source, causing a rapid rise in surface temperature followed by a decay. Due to the rapid diffusion of heat within the material, the flash temperature field is initially sensitive to shallow defects, but the signal from deep defects weakens significantly over time. Therefore, flash thermal imaging is more suitable for the rapid detection of surface and near-surface defects. The main advantages of flash excitation are that a complete time decay curve can be obtained in a single experiment, the function analysis is relatively simple, and there are numerous time-frequency domain data processing methods for thermal image sequences, making it suitable for transient reconstruction analyses such as thermal tomography. However, its detection depth is limited by the material's thermal diffusivity, the duration of the thermal excitation pulse, and the excitation energy. For the detection of deep defects, especially materials with low thermal diffusivity, the thermal response signal of deep defects is weak and easily obscured by surface noise, limiting the depth detection range.
[0004] Modulated thermal imaging (MTI) generates thermal wave signals within materials through periodic thermal excitation. When encountering defects such as delamination or debonding, differences in thermal diffusivity cause phase delays or amplitude attenuation of the thermal waves. The surface temperature response sequence signal is processed to extract amplitude and phase images. Defect characterization is achieved based on the phase difference or amplitude difference between the defect and the substrate, especially the phase image, which is more sensitive to internal material features than the amplitude image and is less susceptible to interference from factors such as uneven heating and material surface conditions. MTI can compensate for the shortcomings of flash mode in deep detection to some extent, but it still faces some challenges in practical applications. The diffusion depth of the modulated thermal wave is closely related to the modulation frequency. Low-frequency thermal waves have strong penetrating power, which can increase the detection depth, but this often leads to severe blurring of shallow feature signals. While increasing the modulation frequency can enhance the resolution of shallow defects, it limits the detection depth. Furthermore, MTI suffers from the "blind frequency" problem in practical applications, where the phase difference or amplitude difference between the defect and the background tends to weaken at specific modulation frequencies, making defects at certain depths difficult to detect. Due to these limitations, it is usually necessary to perform multiple experiments on the same sample at different modulation frequencies to balance detection depth and resolution. However, this multi-frequency experimental method not only increases experimental complexity but also increases detection time, severely reducing detection efficiency. Therefore, how to reduce the impact of blind frequencies, decrease the number and time of modulation thermal imaging tests, and improve depth detection capability and data utilization without sacrificing detection depth and resolution remains an important technical problem to be solved in this field.
[0005] Based on the above, the applicant has identified the following technical problems with the existing technology: First, although flash thermal imaging can obtain a complete transient temperature response with a single excitation, its sensitivity to deep defects is limited, thus restricting its depth imaging capability. Second, modulated thermal imaging can use thermal waves of different frequencies to achieve deep layer detection, but blind frequency phenomenon is prone to occur at specific frequencies, which leads to a decrease or disappearance of defect contrast. Third, in order to cover different depth ranges, traditional modulated thermal imaging requires multi-frequency experiments, which are complex, time-consuming, and difficult to guarantee data consistency. Summary of the Invention
[0006] I. Technical problems to be solved The present invention aims to at least partially solve one of the above-mentioned technical problems.
[0007] II. Technical Solution The first aspect of this invention provides a method for converting modulated thermal imaging data and flash thermal imaging data. This method includes: Step S1: Obtain the temperature-time series of the average temperature of the sample under modulated thermal excitation as a function of time.T ω ( t ),in, To modulate the modulation angular frequency of the thermal excitation; Step S2, temperature time series T ω ( t (as a temperature time series) The initial phase shift of the modulated thermal excitation is obtained by nonlinear fitting using the following formula. ; in, and They are The non-oscillatory components and the oscillatory components, k The thermal conductivity of the sample, The thermal diffusivity of the sample is . L The thickness of the sample; F The heating constant is used to modulate the thermal excitation; ; Step S3, utilizing the initial phase shift of the modulated thermal excitation. Calculation parameters and ; , (14) Step S4: For the temperature time series of a single pixel of the sample under the modulation thermal excitation method. The temperature-time data under flash-like thermal excitation are obtained using the following formula. ( ); (13).
[0008] In some embodiments of the present invention, step S1 includes: sub-step S1a, obtaining a temperature-time series of the temperature change of each pixel of the sample over time under modulation thermal excitation. T ω ( t Sub-step S1b: Average the temperature of all pixels at a given time point to obtain a temperature-time series showing the change of the average sample temperature over time. T ω ( t ).
[0009] In some embodiments of the present invention, in sub-step S1a, the modulated thermal excitation is one of the following two forms: cosine modulation excitation, the modulation heating intensity of which is: The general cosine modulation excitation, which includes a steady-state component and an oscillating component with arbitrary phase angle, has the following modulation heating intensity: .
[0010] In some embodiments of the present invention, before step S1, the method further includes: step S0, which involves periodically heating the sample for a preset period T and a preset duration t0 using a modulated thermal excitation method. During the periodic heating process, thermal image sequences of the sample surface are acquired at a preset acquisition frequency, and a temperature-time sequence of temperature change of each pixel over time is extracted. T ω ( t The method includes at least one of the following: ① cleaning the sample surface; ② covering the sample surface with a water-soluble black coating; ③ using a halogen lamp as a thermal excitation source and driving a silicon controlled rectifier dimmer to generate cosine or sine modulated photothermal energy to periodically heat the sample surface; ④ using an infrared thermal imager to record the change in sample surface temperature over time; ⑤ setting an infrared filter between the heat source and the sample.
[0011] In some embodiments of the present invention, the sample is a flat plate sample or a tubular sample.
[0012] A second aspect of the present invention provides a method for reconstructing modulation signals from thermal imaging data. This method for reconstructing modulation signals from thermal imaging data includes: Perform steps S1 to S4 in the above method for converting modulated thermal imaging data to flash thermal imaging data to obtain temperature-time data under flash-like thermal excitation. ( ); Step S5, using temperature data from flash-like thermal excitation. ( The target modulation frequency is reconstructed using the following formula. Modulated temperature signal ; .
[0013] In some embodiments of the present invention, step S3 is followed by: targeting N different target modulation frequencies. Execute steps S4 to S5 respectively to generate a complete multi-frequency modulated temperature signal, where N ≥ 2.
[0014] In some embodiments of the present invention, step S5 is followed by step S6, which involves reconstructing the modulated temperature signal. By continuously adjusting the fitting parameters through iterative methods, the fitting function is optimized. Corresponding theoretical value and modulated temperature signal The deviation between them gradually decreases, and the optimal fitting coefficients are obtained. and Then, the phase of the thermal wave response signal is calculated. With amplitude ; (19) , .
[0015] In some embodiments of the present invention, steps S4 to S6 are performed on all pixels on the sample to obtain frequency domain feature display: phase map and / or amplitude map, thereby obtaining defect information in the sample.
[0016] A third aspect of the present invention provides a computer device. The computer device includes: a memory and a processor; and a computer program stored in the memory; wherein the processor executes the computer program to implement: the above-described method for converting modulated thermal imaging data to digital flash thermal imaging data; or, the above-described method for reconstructing modulated signals from thermal imaging data.
[0017] III. Beneficial Effects As can be seen from the above technical solution, the present invention has at least one of the following beneficial effects compared to the prior art: (1) This invention solves the problem of time consumption caused by multiple tests in modulation thermal imaging technology. By establishing a conversion model between modulation thermal imaging and flash thermal imaging data, a theoretical basis is provided for the combination of the two technologies.
[0018] (2) The present invention can optimize the detection efficiency of modulated thermal imaging technology by using the modulation signal reconstruction method of thermal imaging data, while taking into account the balance between detection depth and resolution. Attached Figure Description
[0019] Figure 1 This is a schematic diagram of the modulated excitation thermal imaging experimental system.
[0020] Figure 2 This is a schematic diagram of the structure of a GFRP (glass fiber reinforced plastic) flat-bottomed hole specimen.
[0021] Figure 3 This is a schematic diagram of the structure of a stainless steel flat-bottomed hole specimen.
[0022] Figure 4 This is a schematic diagram of the structure of a flash-excited thermal imaging experimental system.
[0023] Figure 5 The measured average modulation temperature is obtained by fitting the theoretical model function using the transformation formula.
[0024] Figure 6 for , , And the change of the converted flash temperature T with time t.
[0025] Figure 7 The change of flash temperature T with time t after log-log coordinate downconversion.
[0026] Figure 8 This is the logarithmic temperature-logarithmic time curve obtained from a flash thermal imaging experiment.
[0027] Figure 9 The graph shows the measured modulation temperature data (4.5 cycles) and the modulation temperature data after inverse conversion (1 cycle) of the GFRP specimen.
[0028] Figure 10 The graph shows the measured modulation temperature data (4.5 cycles) and the modulation temperature data after reverse conversion (1, 2, 4, and 6 cycles) of the GFRP specimen.
[0029] Figure 11 The phase and amplitude diagrams of the modulation cycle data for one cycle were obtained by converting the 4.5-cycle modulation experiment in the forward direction to flash data and then in the reverse direction to reconstruct the data. The trend of the results is the same as that of the phase and amplitude diagrams obtained from the 1-cycle modulation data of the modulation excitation thermal imaging experiment under the same conditions.
[0030] Figure 12 The phase and amplitude diagrams of the modulation temperature (4.5 cycles) measured in the experiment on the GFRP specimen were converted into the phase and amplitude diagrams of the modulation temperature under different modulation cycles.
[0031] Figure 13 The phase and amplitude diagrams of the modulation temperature (6 cycles) measured in the experiment on stainless steel specimens were converted into phase and amplitude diagrams of the modulation temperature under different modulation cycles. Detailed Implementation
[0032] The technical problem of this invention is: how to achieve bidirectional conversion between modulated thermal imaging and flash thermal imaging data based on single-shot modulated thermal imaging experimental data, and further construct a modulated temperature response of arbitrary frequency, so as to reduce the need for multiple experiments, improve the ability to reveal deep defects, and avoid missed detections caused by blind frequency.
[0033] To address the aforementioned technical problems, this invention proposes a data conversion method based on a heat conduction model. This method reconstructs the flash temperature curve using modulation temperature data and then reconstructs the steady-state thermal wave response at any modulation frequency using the flash temperature.
[0034] To facilitate a better understanding of the features of this invention, an explanation will be given from a theoretical perspective combined with examples. There is a direct relationship between the surface temperature measured in flash thermal imaging tests and the surface temperature in modulated thermal imaging tests. To illustrate this relationship, the applicant considered the ideal Dirac... Theoretical surface temperature obtained in flash thermal imaging under pulse heating ( ), and based on Duhamel's principle from ( ) Derived surface temperature in modulated thermal imaging ( ). The following will derive this relationship under two different modulation excitation sources: one is an excitation source containing only oscillating components, and the other is an excitation source containing steady-state components and oscillating components with arbitrary excitation phase angles.
[0035] In thermal imaging, the excitation signal is modulated. It typically changes periodically, assuming the modulating signal... Containing only oscillatory components, it can be expressed as: (1) in F Let be a constant, and take in the following derivation. F = 1, = 2π f Angular frequency, f The modulation frequency is given in Hz, and the resulting solution for the modulation surface temperature is: (2) Taking the first and second time derivatives of the above equation, we get: (3) (4) The integral of equation (4) is the same as that in equation (2). Therefore, substituting equation (2) into equation (4) yields: (5) Integrating the above equation over time, we get: (6) Thermal excitation When the modulated excitation source containing only oscillatory components is described by equation (1), equations (2) and (6) can be used to extract flash excitation data (temperature solution). Convert to modulation data (temperature solution) ,vice versa.
[0036] If the excitation source is a general cosine-modulated component containing a steady-state component and an oscillating component with an arbitrary phase angle, the modulation heating intensity... Defined as: (7) Among them, take Given the initial phase offset value in radians, the solution for the modulation surface temperature is: (8) Taking the first and second time derivatives of the above equation, we get: (9) Since the integral component in equation (10) is the same as that in equation (8) Therefore, we get: Taking the time derivative of the above equation, we get: Equation (12) is about The second-order nonhomogeneous ordinary differential equation, after derivation and calculation, can finally be obtained. The solution is: (13) in, , (14) If the initial phase shift is determined and modulation frequency The value, parameter and If all can be determined, then equation (8) can be used to solve for the flash temperature. Convert to modulated temperature solution Equation (13) can be used to decode the modulation temperature. Convert to flash temperature solution The premise is that the thermal excitation method It is given by equation (7). Since no special conditions are specified in the derivation, this transformation is applicable to various cases with or without surface convection and material translucency.
[0037] Experiments were conducted to verify the conversion between modulated thermal imaging and flash thermal imaging data on flat-bottomed hole specimens made of two existing materials. Figure 2 and Figure 3These are fiberglass reinforced plastic (GFRP) specimens and stainless steel specimens. First, the GFRP specimens were thermally excited using a halogen lamp cosine heating method. This thermal excitation was achieved by simulating a cosine function driving the halogen lamp through a step function with different initial phase shifts. The following simulation uses a modulation function with a modulation period of 33 seconds. , f =1 / 33=0.03Hz =2 =0.19 s 1 Although the applicant's simulated curve has local waveform differences from the ideal cosine curve, it does not affect the conversion of the thermal signal. In subsequent data processing, it will be corrected by smoothing and least squares fitting methods.
[0038] In order to obtain accurate data conversion results, the applicant, in addition to the accurate modulation frequency provided by the dimmer, also needs... It is also necessary to determine the accurate initial phase shift. This can be achieved by fitting experimental data with theoretical solutions. The theoretical model assumes that the thickness of the tested flat plate sample is... L to its surface Modulated thermal excitation, in arbitrary At this value, the surface temperature The general theoretical solution is: in, and They are The non-oscillatory components and the oscillatory components, k Thermal conductivity, Let be the thermal diffusivity, and ,Will and By fitting the surface temperature data obtained from the modulated thermal imaging experiment, an accurate initial phase shift can be obtained. .
[0039] In the known and In the case of values, the parameters in equation (13) and The modulation temperature can be determined according to equation (14) and can be obtained through equation (13). ( Forward conversion to flash temperature ( The right side of the forward conversion formula (13) contains three terms: modulation temperature. Its time derivative and convolution terms These three parts respectively reflect the temperature response of the surface of the object being tested after the modulation frequency heat flow is input, the transient rate of change of surface temperature, and the historical thermal response characteristics of the modulation temperature. Figure 6 In, it is shown , / , and the converted flash temperature How it changes over time. From Figure 6 As can be seen, the first three components all exhibit obvious periodic oscillations in the original time domain, which is caused by the modulation excitation itself, and the amplitudes of the three oscillation components are all much greater than the converted flash temperature. T However, based on the calculations of this formula, through precise phase and amplitude matching, the periodic changes of these three large-amplitude oscillation components will mathematically cancel each other out, ultimately outputting a stable, non-oscillating temperature signal. T The applicant will convert the flash temperature. T The changes over time t are all logarithmic and replotted on logarithmic coordinates to more clearly observe the pattern of temperature change, such as... Figure 7 As shown, the reconstructed flash temperature time series curve in the logarithmic coordinate system exhibits a significant time-dependent decay characteristic.
[0040] For a semi-infinite medium model, the surface temperature changes with time under flash lamp excitation as follows: (16) when L As it approaches infinity, it follows the following: (17) When taking log-log coordinates (horizontal axis) , vertical axis The above formula is transformed to obtain: (18) This will result in a slope of The straight line, from Figure 7 The results of the observed data transformation in the logarithmic coordinate system conform to the characteristics of the surface temperature-time curve under flash excitation, and also have approximate... The slope.
[0041] At the same time, the applicant used Figure 4 The experimental setup was used to conduct a real flash-excited thermal imaging experiment on the same specimen. As shown in the figure, under the control of a computer (PC) and a control unit, two flash lamps heated the specimen, and an infrared thermal imager (IR camera) collected infrared thermal data, which was then transmitted back to the computer (PC) for processing.
[0042] This method yielded actual temperature data from the flash experiment, and the data underwent the same processing, such as... Figure 8 As shown. By observation Figure 7 and Figure 8 It can be seen that the flash temperature data obtained by converting the modulated temperature data is highly consistent with the flash temperature data measured in the actual flash thermal imaging experiment, which verifies that the modulated thermal wave signal contains all the physical information sufficient to reconstruct the flash signal response. In summary, the applicant has verified the correctness and feasibility of the forward conversion from modulated thermal imaging data to flash thermal imaging data.
[0043] Flash temperature profile obtained by forward conversion As input, the modulated temperature signal at any modulation frequency was further reconstructed using the inverse transformation formula (8). The system verified the feasibility and accuracy of the conversion mechanism from forward conversion of modulation data to flash data and then back to modulation data. Although there are significant differences between the surface thermal images generated by various excitation forms (such as flash and modulation), since the basic thermal processes of these forms of excitation are the same—that is, they all involve transient heat transfer into the material—there must be some correlation between the surface temperature data and the processing results under these forms of excitation. This reconstruction is essentially a reorganization of complete information, rather than a reduction or truncation of the data. Therefore, the forward and reverse conversion does not cause any loss of physical information, and theoretically, it should be possible to obtain a steady-state thermal wave signal equivalent to that of a real modulation experiment. To verify this characteristic, Figure 9 The image shows a single pixel from the center of the image, comparing the surface temperature changes measured in the original image (4.5 cycles, acquisition time 150s, modulation period 33s, modulation frequency 0.03Hz, modulation angular frequency). =2 =0.19 s 1 The single-cycle modulation angular frequency obtained by the inverse conversion is: =0.042 s 1 The reconstructed temperature curve (with a modulation frequency of 0.0067Hz, corresponding to only one complete cycle within a total duration of 150 seconds) is shown in the figure, and the corresponding fitted curve is also superimposed on it.
[0044] Experimental results are as follows Figure 9 and Figure 10 By comparing the data of the same central pixel, it can be seen that the different modulation period data (periods 1, 2, 4, and 6) generated by the reverse reconstruction all show the same overall trend and waveform characteristics as the measured 4.5-period modulation experiment, proving the reliability of the frequency domain reconstruction method. In particular, the phase delay and amplitude images extracted by the fitting method (FIT) are shown in the following formula: in, 0、 , , The constant is the fitting constant. The first two terms on the right-hand side of the formula are non-oscillatory components, and the last two terms are oscillatory components. By iteratively adjusting the fitting parameters, the deviation between the theoretical value and the experimental data corresponding to the fitting function is gradually reduced, eventually yielding the optimal fitting value. The fitting coefficients are used to... and The phase of the thermal wave response signal can be directly calculated according to equation (21). With amplitude : , The phase and amplitude diagrams of the modulation period data (one cycle) generated by the 4.5-cycle modulation experiment, after being converted from forward to flash data and then reconstructed from reverse, are highly consistent with the phase and amplitude diagrams obtained from the 1-cycle modulation data of the modulation-excited thermal imaging experiment under the same conditions. Figure 11 As shown, this demonstrates that the inverse conversion can not only reproduce the equivalent modulation signal in the time domain, but also maintain strict consistency with the real experiment in the detection of key detection parameters.
[0045] Table 1 Further processing of each pixel in the full frame image using the same forward and reverse methods yields the phase and amplitude images, as shown below. Figure 12 As shown. (Through) Figure 12It can be seen that the multi-frequency modulation data obtained by the reverse conversion of this invention can accurately reflect the contrast variation law of defects of different depths at different frequencies, including the typical thermal wave behavior of shallow defects being significant at high frequencies and deep defects being clearest at low frequencies. Regarding the phase parameter, when the excitation frequency is lower than the blind frequency, the defect area exhibits a brighter phase delay (positive phase delay); while when the excitation frequency is higher than the blind frequency, the brightness decreases or even reverses (negative phase delay). Similar patterns were observed in amplitude analysis: under low-frequency excitation, the defect amplitude is significantly higher than the background (bright area), while under high-frequency excitation, the amplitude decays (dark area). Furthermore, the amplitude is less sensitive to defect depth; shallow defects are only significant at high frequencies. Deep defects require low-frequency signals to trigger sufficient thermal wave penetration, but the amplitude response is weak, requiring phase analysis. This frequency-dependent contrast variation provides a key criterion for defect detection; that is, by selecting an appropriate modulation frequency, the identifiability of defects in thermal imaging can be effectively enhanced. This is not only strictly consistent with the thermal diffusion theory and the blind frequency model (Table 1), but also proves the physical correctness of the reverse conversion method in the defect manifestation mechanism.
[0046] The above results show that the forward and reverse data conversion processing of Equations (13) and (8) can be flexibly converted into an equivalent modulation signal of any frequency. This process does not require repeated experiments, and only one modulation test data (4.5 cycles) is needed to realize the conversion of multi-frequency detection results (reconstructing any cycle such as 1, 2, 4, 6 cycles), which solves the bottleneck of traditional methods that require multiple experiments.
[0047] In addition to the results mentioned above, modulation-excited thermal imaging experiments were also conducted on stainless steel flat-bottomed hole plate samples. Similar results were obtained by converting the modulation data forward to flash data and then inversely converting it to multi-frequency modulation data, extracting the phase and amplitude diagrams, such as... Figure 13 As shown, the applicability of this conversion method to both non-metallic and metallic materials is verified.
[0048] Based on the above theoretical explanation, this invention provides a method for converting modulated thermal imaging data and flash thermal imaging data, solving the problem of time consumption caused by multiple tests in modulated thermal imaging technology. By establishing a conversion model between modulated thermal imaging and flash thermal imaging data, a theoretical basis is provided for the combination of the two technologies, and an efficient data processing method is provided to optimize the detection efficiency of modulated thermal imaging technology while balancing detection depth and resolution.
[0049] In an exemplary embodiment of the present invention, a method for converting modulated thermal imaging data and flash thermal imaging data is provided. The method for converting modulated thermal imaging data and flash thermal imaging data in this embodiment includes: Step S0: The sample is periodically heated for a preset period T and a preset duration t0 using a modulated thermal excitation method. During the periodic heating process, thermal image sequences of the sample surface are acquired at a preset acquisition frequency, and the temperature-time sequence of each pixel is extracted. T ω ( t ); Figure 1 This is a schematic diagram of the structure of a modulation-excited thermal imaging experimental system. Figure 1 As shown, in this embodiment, two halogen lamps are used as thermal excitation sources. A digital triac dimmer is used to drive the generation of cosine or sine modulated photothermal energy to periodically heat the sample surface. An infrared thermal imager (IR camera) records the temperature change of the sample surface over time, and the acquired signals are transmitted to a computer for processing. Defects within the sample exhibit a phase delay relative to defect-free areas, resulting in different grayscale levels, which allows for defect detection.
[0050] The following aspects should be noted in this step: ① Adjustment of the thermal imager Before the experiment began, the focus of the thermal imager was adjusted and non-uniform correction was performed to ensure the uniformity of the temperature response in the field of view of the thermal imager and to correct the corresponding differences of each pixel of the detector.
[0051] ② Sample cleaning and surface treatment Before the experiment begins, the sample surface needs to be cleaned to remove dirt to avoid interference with the test data. After it dries naturally, it should be covered with a water-soluble black coating, which can improve the surface emissivity of the material and enhance the photothermal conversion performance.
[0052] ③ Add an infrared filter device Since heat sources such as halogen lamps and flash lamps emit infrared radiation, an infrared filter is installed in front of the lamp to prevent the infrared radiation from the excitation source from being directly reflected to the infrared camera, which would seriously interfere with the acquisition of thermal radiation signals from the sample surface.
[0053] ④ Modulation of the thermal excitation period The surface of the object under test is periodically heated with a sine or cosine wave for a preset duration (e.g., 150s) and a period T (e.g., T=33s) using a modulation / phase-locked thermal excitation device. This process should include at least one complete cycle. The thermal image sequence of the heating process of the object surface is then acquired by an infrared thermal imager at a preset acquisition frequency. The acquired surface thermal image sequence is then transmitted and stored in the general-purpose memory of a computing device.
[0054] ⑤ Sample shape In this embodiment, the sample can be a flat plate or a tubular sample. However, the invention is not limited to this. In other embodiments of the invention, the sample can also be a blocky or irregularly shaped sample, and the invention can still be used to detect internal defects.
[0055] Step S1: Obtain the temperature-time series of the average temperature of the sample under modulated thermal excitation as a function of time. T ω ( t ),in, To modulate the modulation angular frequency of the thermal excitation; It should be noted that the modulation angular frequency of the modulated thermal excitation... This is related to the pre-set period T. In this embodiment, T = 33s, therefore, f =0.03Hz, =2 =0.19 s 1 Regarding the preset period, its value range is very wide, ranging from milliseconds to one day, one month, or even one year. Therefore, the modulation angular frequency... The range of values for is also very wide, which will not be elaborated here.
[0056] Further, step S1 includes: Sub-step S1a obtains the temperature-time series of the temperature change of each pixel of the sample under modulation thermal excitation over time. T ω ( t ); Preferably, the modulated thermal excitation takes one of the following two forms: ① Cosine modulation excitation, the modulation heating intensity is: ; ② A general cosine modulation excitation containing a steady-state component and an oscillating component with arbitrary phase angles has the following modulation heating intensity: .
[0057] Those skilled in the art should understand that the above two modulation thermal excitation methods are easy to implement and the data processing is relatively simple. Besides the two methods mentioned above, other modulation thermal excitation methods can also be used, such as sinusoidal modulation excitation, sawtooth wave modulation excitation, and rectangular wave modulation excitation, all of which can achieve the present invention.
[0058] Sub-step S1b involves averaging the temperature of all pixels at a given time point to obtain a temperature-time series showing the change of the average sample temperature over time. T ω ( t ).
[0059] Step S2, temperature time series T ω ( t (as a temperature time series) The initial phase shift of the modulated thermal excitation is obtained by nonlinear fitting using the following formula. ; in, and They are The non-oscillatory components and the oscillatory components, k The thermal conductivity of the sample, The thermal diffusivity of the sample is . L The thickness of the sample; F The heating constant is used to modulate the thermal excitation; ; Step S3, utilizing the initial phase shift of the modulated thermal excitation. Calculation parameters and ; , (14) Step S4: For the temperature time series of a single pixel of the sample under the modulation thermal excitation method. The temperature-time data under flash-like thermal excitation are obtained using the following formula. ( ); (13) It should be noted that the above steps use the term "flash-like thermal excitation" mainly to indicate that the temperature and time data are not actual flash thermal imaging data, but data converted from modulated thermal imaging.
[0060] As can be seen, the above operations achieve the conversion from modulated thermal imaging data to flash thermal imaging data, thus laying the foundation for subsequently combining the advantages of the two thermal imaging methods. Of course, this invention does not preclude performing other types of operations after converting the thermal imaging data into flash-like thermal imaging data, which will not be elaborated here.
[0061] This concludes the description of the method for converting modulated thermal imaging data and flash thermal imaging data according to embodiments of the present invention.
[0062] Based on the above-described method for converting modulated thermal imaging data to flash thermal imaging data, a second aspect of the present invention further provides a method for reconstructing the modulated signal of thermal imaging data. In an exemplary embodiment of the present invention, the method for reconstructing the modulated signal of thermal imaging data includes: First, steps S1 to S4 of the method for converting modulated thermal imaging data to flash thermal imaging data in the above embodiment are performed to obtain temperature-time data under flash-like thermal excitation. ( ); Step S5, using temperature data from flash-like thermal excitation. ( The target modulation frequency is reconstructed using the following formula. Modulated temperature signal ; (8) It should be noted that, in other embodiments of the present invention, after step S3, the method further includes: targeting N different target modulation frequencies. By executing steps S4 to S5 respectively, a complete multi-frequency modulated temperature signal can be generated, where N ≥ 1.
[0063] Specifically, the flash temperature curve obtained by forward conversion As input, the inverse conversion formula (8) can be used to reconstruct any set modulation frequency (e.g. =0.042 s 1 or f Modulated temperature signal at 0.0067Hz It can modulate multiple target frequencies (e.g.) =0.084 s 1 or f =0.013Hz, =0.167 s 1 or f =0.027Hz, =0.251 s 1 or f Repeat the above process (=0.04Hz) to generate a complete multi-frequency modulated temperature signal.
[0064] As can be seen from the above description, this embodiment uses a known thermal diffusivity. The modulation frequency obtained by fitting the sample with formulas (15, 16) and initial phase shift A forward conversion model can be constructed, and the equivalent flash temperature curve can be obtained through the forward conversion formula (13). The transient temperature reconstructed based on the forward conversion... T (t), through the reverse conversion of formula (8), a modulated temperature signal at any frequency is generated to realize the reconstruction of multi-frequency modulated data.
[0065] Step S6, for the reconstructed modulated temperature signal By continuously adjusting the fitting parameters through iterative methods, the fitting function is optimized. Corresponding theoretical value and modulated temperature signal The deviation between them gradually decreases, and the optimal fitting coefficients are obtained. and Then, the phase of the thermal wave response signal is calculated. With amplitude ; (19) , Step S7: Perform steps S4-S6 for all pixels on the sample to obtain frequency domain feature display: phase map and / or amplitude map, such as... Figure 11 As shown.
[0066] After obtaining the phase map and / or amplitude map, the sensitivity difference of different frequencies to deep and shallow defects is then utilized to achieve the effect of low-frequency manifestation of deep defects and high-frequency enhancement of shallow defects.
[0067] This concludes the description of the modulation signal reconstruction method for thermal imaging data according to embodiments of the present invention.
[0068] Based on the above embodiments of the method for converting modulated thermal imaging data to flash thermal imaging data and the method for reconstructing modulated signals of thermal imaging data, a third aspect of the present invention also provides a computer device. In an exemplary embodiment of the present invention, the computer device includes: Memory and processor; and, Computer programs stored in memory; The processor executes a computer program to implement: the method for converting modulated thermal imaging and data flash thermal imaging data as described in the above embodiment; or the method for reconstructing the modulated signal of thermal imaging data as described in the above embodiment.
[0069] This concludes the description of the various embodiments of the present invention. Based on the above description, those skilled in the art should have a clear understanding of the present invention.
[0070] It should be noted that for some implementation methods, if they are not the key content of this invention and are well known to those skilled in the art, they are not described in detail in the accompanying drawings or text due to space limitations. In such cases, they can be understood by referring to the relevant prior art.
[0071] Unless explicitly stated otherwise, the numerical values and ranges mentioned in this invention are approximate and can be changed according to the content of this invention. Specifically, all figures in the specification and claims indicating the content of composition, reaction conditions, etc., should be understood to be modified by the term "about" in all cases, meaning that they include variations of ±10% in certain embodiments.
[0072] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by the computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.
[0073] The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such a program implementing the invention can be stored on a computer-readable medium or can take the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
[0074] This invention can be implemented using hardware comprising several different components and a suitably programmed computer. Various component embodiments of the invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. The physical implementation of the hardware structure includes, but is not limited to, physical devices, including, but not limited to, transistors, memristors, DNA computers, microcontrollers, microprocessors, or digital signal processors (DSPs). Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of this invention can be implemented using various programming languages; the description of specific languages herein is for the purpose of disclosing the best mode of implementation of the invention.
[0075] Furthermore, the above embodiments are provided only to enable the invention to meet legal requirements, and the invention can be implemented in many different forms and should not be construed as limited to the embodiments set forth herein.
[0076] Similarly, it should be understood that, for the sake of brevity, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, this method of invention should not be construed as reflecting an intention that the claimed invention requires more features than expressly recited in each claim. Rather, as reflected in the claims, the various inventive aspects consist of fewer than all the features of the preceding single embodiment. Furthermore, embodiments may be used in combination with each other or with other embodiments based on design and reliability considerations; that is, technical features from different embodiments can be freely combined to form more embodiments. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.
[0077] The above specific embodiments have provided a detailed description of the purpose, technical means, and beneficial effects of the present invention. It should be understood that the purpose of the detailed description is to enable those skilled in the art to better understand the present invention, and it 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 converting modulated thermal imaging data and flash thermal imaging data, characterized in that, include: Step S1: Obtain the temperature-time series of the average temperature of the sample under modulated thermal excitation as a function of time. T ω ( t ),in, The modulation angular frequency for modulating thermal excitation; Step S2, the temperature time series T ω ( t ) as a temperature time series The initial phase shift of the modulated thermal excitation is obtained by nonlinear fitting using the following formula. ; in, and They are The non-oscillatory components and the oscillatory components, k The thermal conductivity of the sample, The thermal diffusivity of the sample is . L The thickness of the sample; F The heating constant is used to modulate the thermal excitation; ; Step S3, utilizing the initial phase shift of the modulated thermal excitation. Calculation parameters and ; ， （14） Step S4: For the temperature time series of a single pixel of the sample under the modulation thermal excitation method. The temperature-time data under flash-like thermal excitation are obtained using the following formula. ( ); （13）。 2. The method for converting modulated thermal imaging data and flash thermal imaging data according to claim 1, characterized in that, Step S1 includes: Sub-step S1a obtains the temperature-time series of the temperature change of each pixel of the sample under modulated thermal excitation over time. T ω ( t ); Sub-step S1b involves averaging the temperature of all pixels at a given time point to obtain a temperature-time series showing the change of the average sample temperature over time. T ω ( t ).
3. The method for converting modulated thermal imaging data and flash thermal imaging data according to claim 2, characterized in that, In sub-step S1a, the modulated thermal excitation takes one of the following two forms: The cosine modulation excitation has the following modulation heating intensity: ; The general cosine modulation excitation, which includes a steady-state component and an oscillating component with arbitrary phase angle, has the following modulation heating intensity: .
4. The method for converting modulated thermal imaging data and flash thermal imaging data according to claim 2, characterized in that, The procedure preceding step S1 also includes: Step S0: The sample is periodically heated for a preset period T and a preset duration t0 using a modulated thermal excitation method. During the periodic heating process, thermal image sequences of the sample surface are acquired at a preset acquisition frequency, and the temperature-time sequence of each pixel is extracted. T ω ( t ), which includes at least one of the following: ① Clean the sample surface; ② Cover the sample surface with a water-soluble black coating; ③ Use a halogen lamp as a thermal excitation source, and use a silicon controlled rectifier to drive the generation of cosine or sine modulated photothermal to periodically heat the sample surface. ④ Use an infrared thermal imager to record the change in sample surface temperature over time; ⑤ Install an infrared filter between the heat source and the sample.
5. The method for converting modulated thermal imaging data and flash thermal imaging data according to any one of claims 1 to 4, characterized in that, The sample is either a flat plate sample or a tubular sample.
6. A method for reconstructing modulated signals from thermal imaging data, characterized in that, include: Performing steps S1 to S4 in the method for converting modulated thermal imaging data to flash thermal imaging data as described in any one of claims 1 to 5, temperature-time data under flash-like thermal excitation is obtained. ( ); Step S5, using temperature data from flash-like thermal excitation. ( The target modulation frequency is reconstructed using the following formula. Modulated temperature signal ; .
7. The method for reconstructing modulated signals from thermal imaging data according to claim 6, characterized in that, The step S3 is followed by: For N different target modulation frequencies Execute steps S4 to S5 respectively to generate a complete multi-frequency modulated temperature signal, where N ≥ 2.
8. The method for reconstructing modulated signals from thermal imaging data according to claim 7, characterized in that, Following step S5, the following is also included: Step S6, for the reconstructed modulated temperature signal By continuously adjusting the fitting parameters through iterative methods, the fitting function is optimized. Corresponding theoretical value and modulated temperature signal The deviation between them gradually decreases, and the optimal fitting coefficients are obtained. and Then, the phase of the thermal wave response signal is calculated. With amplitude ; （19） , 。 9. The method for reconstructing modulated signals from thermal imaging data according to claim 8, characterized in that, Steps S4 to S6 are performed on all pixels on the sample to obtain frequency domain feature display: phase map and / or amplitude map, thereby obtaining defect information in the sample.
10. A computer device, characterized in that, include: Memory and processor; as well as, Computer programs stored in memory; The processor executes the computer program to implement: a method for converting modulated thermal imaging and data flash thermal imaging data as described in any one of claims 1 to 5; or a method for reconstructing modulated signals of thermal imaging data as described in any one of claims 6 to 9.