Blind structure polarization modulation infrared thermal wave super-resolution imaging detection system and method

Abstract

The application belongs to the technical field of infrared thermal wave imaging, and particularly relates to a blind structure polarization modulation infrared thermal wave super-resolution imaging detection system and method, which comprises an infrared camera, a processor, a signal generator, a power amplifier, a semiconductor laser and a structured light modulator. The system and method reconstruct a super-resolution detection image through an instantaneous phase difference image of a thermal wave signal and a linear frequency modulation signal under different spatial polarization phases, and improve the spatial resolution of infrared thermal wave imaging in detecting tiny defects and complex edge defects.

Description

Technical Field

[0001] This application belongs to the field of infrared thermal imaging technology, specifically relating to a blind structure polarization modulation infrared thermal super-resolution imaging detection system and method. Background Technology

[0002] Infrared thermal imaging inspection technology is an active infrared non-destructive testing method. It typically uses methods such as optical excitation, ultrasonic excitation, and eddy current excitation to heat the test sample. An infrared camera collects the surface temperature change history of the test sample, and by analyzing the time-domain or frequency-domain characteristics of the thermal wave signal, it quantitatively characterizes the differences between defective and non-defective areas. This method has advantages such as high detection efficiency, intuitive and easy-to-understand detection results, wide applicability, and non-contact detection, and has been widely used in aerospace, medical, construction, and semiconductor manufacturing fields.

[0003] Ultrasonic excitation infrared thermography detects defects by exciting internal vibrations in materials with ultrasonic pulses. Defect areas experience localized temperature rises due to friction and plastic deformation. Infrared cameras analyze these temperature changes to detect defects, offering good detection of surface and subsurface cracks. However, the ultrasonic excitation device must be in close contact with the sample, and its effectiveness is limited for materials with poor thermal conductivity. Eddy current infrared thermography excites eddy currents within materials using alternating current. Defect areas affect eddy current distribution and heat conduction characteristics. Infrared cameras analyze surface temperature differences to detect and characterize internal defects. However, this excitation method is only suitable for conductive materials, and edge features in the detected images are not significant. Optically excited infrared thermography is one of the most widely used methods. This method typically uses modulated lasers or halogen optical heat sources to heat the sample. Heat waves are conducted from the surface to the interior of the material, interacting with defects and causing temperature differences on the sample's surface. Infrared cameras record and analyze the time-frequency and characteristics of the heat wave signals to achieve efficient detection of internal defects. This method has no special limitations on the tested materials and has a wide range of applications. However, the spatial resolution of the aforementioned infrared thermal imaging detection methods all depend on the spatial resolution of the infrared camera. However, due to the limitations of semiconductor material properties, the pixel size of the infrared detector is relatively large, making it difficult to further improve the spatial resolution of the infrared thermal imaging detection system. This results in insufficient detection capabilities for minute defects and complex edge defects, making it difficult to provide accurate defect damage characteristic information for structural safety evaluation. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a blind structure polarization modulation infrared thermal super-resolution imaging detection system and method. It employs pulse compression time-domain modulation, structured light spatial modulation, and polarization light modulation techniques to reconstruct super-resolution detection images from the instantaneous phase difference images of thermal wave signals and linear frequency modulation signals under different spatial polarization phases, thereby improving the spatial resolution of infrared thermal wave imaging for detecting minute defects and complex edge defects.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A blind structured polarization modulated infrared thermal super-resolution imaging detection system includes an infrared camera, a processor, a signal generator, a power amplifier, a semiconductor laser, and a structured light modulator. The infrared camera is connected to the processor. One analog signal output of the signal generator is connected to the power amplifier, and the other is connected to the structured light modulator. The light spot emitted by the structured light modulator illuminates the test sample. The power amplifier is connected to the semiconductor laser, and the semiconductor laser is connected to the structured light modulator. The infrared camera, signal generator, and structured light modulator are each connected to the processor. The structured light modulator includes a lens sleeve and a collimating lens, an engineered diffuser, a polarizing mirror, and a stepper motor mounted on the sleeve. The collimating lens is fixed to the input end of the lens sleeve via a second adapter ring. The engineered diffuser is fixed inside the lens sleeve via an internal thread. The polarizing mirror is fixed to the output end of the lens sleeve via a first adapter ring. The collimating lens is connected to a semiconductor laser via an optical fiber, and the stepper motor is connected to a signal generator.

[0006] Preferably, the infrared camera operates at a wavelength of 3µm to 5µm, has a frame rate of ≥50Hz, an equivalent noise error of <30mK, and a field of view greater than 22°×16°.

[0007] Preferably, the collimating lens is connected to the input end of the lens sleeve via a second adapter ring, with a working center wavelength of 780nm, a numerical aperture NA>0.25, and a focal length f>36 mm.

[0008] Preferably, the engineered diffuser shapes the laser spot after it is aligned by diffuse reflection, and the output spot is an approximately uniform surface heat source with a divergence angle greater than 20° and a wavelength range of 380nm~1100nm with a transmittance greater than 90%.

[0009] Preferably, the polarizing mirror outputs linearly polarized light with a working wavelength of 300nm~3um, an extinction ratio of >600:1, a reflectivity of less than 1% and a transmittance of >85% when incident normally, and is connected to the output end of the lens sleeve through a first adapter ring.

[0010] A blind structure polarization-modulated infrared thermal wave super-resolution imaging detection method is proposed. This method employs a linear frequency modulated pulse compression signal to temporally modulate a semiconductor laser excitation source, while simultaneously using an engineering diffuser and a polarizing mirror to spatially modulate the laser beam. After the test sample absorbs the temporally regulated excitation heat flow, the thermal wave conducts heat within the material. The change in the surface temperature signal of the test sample is recorded by an infrared camera. A Hilbert transform is used to construct an analytical feature signal, extracting the time-frequency domain features of the surface thermal wave signal. A feature image sequence is constructed by analyzing the instantaneous phase difference between defective and non-defective regions. A two-dimensional Fourier transform is used to extract the high-frequency and low-frequency components of the phase detection feature image sequence under different polarization directions. The high-frequency and low-frequency components of the feature image sequence obtained from three detections are fused, and a super-resolution reconstructed image is obtained through a two-dimensional inverse Fourier transform.

[0011] Preferably, the semiconductor laser outputs pulse compression time-domain modulated laser; (1); In the formula, M ( t () is a linear frequency modulated pulse compression signal. V 0 represents the voltage of the modulation signal. f 0 is the initial frequency of linear frequency modulation. k It's frequency adjustment. t It is a moment.

[0012] Preferably, the thermal wave signal from the surface of the test sample is processed, and an analytical characteristic signal is constructed using Hilbert transform. A α ( t Its construction method is as follows: (2); i It is a complex factor, H[.] is the Hilbert transform, ( x , y ) represents the pixel position. α It is the rotation angle. T α ( x , y , t ) represents a specific pixel in an infrared image sequence. t Temperature value at any given time; Instantaneous phase feature information of thermal wave signals is extracted using arctangent operation. The calculation method is as follows: (3); Calculate the instantaneous phase difference between the surface thermal wave signal and the linear frequency modulated pulse compression signal corresponding to each pixel. The phase difference feature is used to construct an infrared thermal wave detection image sequence; (4).

[0013] Preferably, the spatial spectrum of the infrared thermal wave detection phase feature image is extracted using two-dimensional Fourier transform. The high-frequency components are The low-frequency component is ,in, (5); In the formula, w, v Spatial frequency of the image f max The highest spatial frequency of the feature image.

[0014] Preferably, the high-frequency components of the feature image sequence obtained from three detections and low frequency components Pixel-level fusion is performed using the following fusion strategy: (6); In the formula, MAX represents taking the maximum value, and MIN represents taking the minimum value. and Constructing the two-dimensional Fourier frequency of the super-resolution reconstructed image Super-resolution reconstructed images are obtained through 2-bit inverse Fourier transform. Achieving super-resolution reconstruction of infrared thermal wave detection images under blind structure polarization-modulated laser excitation: (7).

[0015] Compared with the prior art, the beneficial effects of this application are as follows: The method of this invention can improve the spatial resolution of infrared thermal imaging detection systems. Compared with traditional structured light illumination super-resolution reconstruction methods, the proposed reconstruction method does not require the intensity distribution of structured light, has a fast detection speed, and improves the reconstruction efficiency of super-resolution images. The system and method disclosed in this invention are simple to operate, have high detection efficiency, and high detection resolution, and have excellent application prospects. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0017] Figure 1 This is a schematic diagram of a blind structure polarization modulation infrared thermal super-resolution imaging detection system disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structured light modulator disclosed in an embodiment of the present invention; Figure 3 For super-resolution image reconstruction; In the diagram, 1. Test sample; 2. Infrared camera; 3. First data line; 4. Processor; 5. First control line; 6. Signal generator; 7. Second control line; 8. Third control line; 9. Power amplifier; 10. Power line; 11. Semiconductor laser; 12. Optical fiber; 13. Structured light modulator; 14. Collimating lens; 15. Lens sleeve; 16. Engineering diffuser; 17. First adapter ring; 18. Polarizing mirror; 19. Transmission gear; 20. Stepper motor; 21. Second adapter ring; 22. Blind structured polarization modulation spot. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0019] A blind structured polarization modulated infrared thermal super-resolution imaging detection system includes an infrared camera 2, a processor 4, a signal generator 6, a power amplifier 9, a semiconductor laser 11, and a structured light modulator 13. The infrared camera 2 is connected to the processor 4. One analog signal output terminal of the signal generator 6 is connected to the power amplifier 9, and the other is connected to the structured light modulator 13. The power amplifier 9 is connected to the semiconductor laser 11, and the semiconductor laser 11 is connected to the structured light modulator 13. The light spot emitted by the structured light modulator 13 illuminates the test sample 1. The infrared camera 2, the signal generator 6, and the structured light modulator 13 are all connected to the processor 4. The system and method disclosed in this invention are simple to operate, have high detection efficiency, and high detection resolution, and have good application prospects.

[0020] In the above scheme, the structured light modulator 13 includes a lens sleeve 15 and a collimating lens 14, an engineered diffuser 16, a polarizer 18, and a stepper motor 20 mounted on the lens sleeve 15. The collimating lens 14 is fixed to the input end of the lens sleeve 15 via a second adapter ring 21. The engineered diffuser 16 is fixed inside the lens sleeve 15 via an internal thread. The polarizer 18 is fixed to the output end of the lens sleeve 15 via a first adapter ring 17. The collimating lens 14 is connected to a semiconductor laser 11 via an optical fiber 12. The stepper motor 20 is connected to a signal generator 6 via a control line.

[0021] In the above scheme, the processor is equipped with a signal generator control module, a thermal wave signal feature extraction module, and a super-resolution image reconstruction module.

[0022] In the above scheme, the infrared camera is used to collect the surface temperature thermal image sequence of the test sample 1, and is connected to the processor through the first data line 3 (Ethernet data line). The working wavelength is 3um~5um, the pixel resolution is better than 320×256, the frame rate is ≥50Hz, the equivalent noise error is <30mK, and the field of view is better than 22°×16°.

[0023] In the above scheme, the signal generator is connected to the processor through the first control line 5 (USB data line), with a frequency resolution of 1mHz and a frequency modulation range of 2mHz~20kHz.

[0024] In the above scheme, the signal generator and the power amplifier are connected through the second control line 7 (BNC data line), the semiconductor laser and the power amplifier are connected through the power supply line, and the power amplifier has an external trigger analog modulation function with a maximum modulation frequency >1kHz.

[0025] In the above scheme, the semiconductor laser is connected to the structured light modulator via optical fiber 12.

[0026] In the above scheme, the signal generator and the structured light modulator are connected via the third control line 8 (BNC data line).

[0027] In the above scheme, the collimating lens is connected to the input end of the lens sleeve through a second adapter ring, with a working center wavelength of 780nm, a numerical aperture NA>0.25, and a focal length f>36mm.

[0028] In the above scheme, the engineering diffuser shapes the laser spot after it is aligned by diffuse reflection. The output spot is an approximately uniform surface heat source with high spatial frequency characteristics, a divergence angle greater than 20°, and a wavelength range of 380nm~1100nm with a transmittance greater than 90%.

[0029] In the above scheme, the polarizing mirror outputs linearly polarized light with a working wavelength of 300nm~3um, an extinction ratio of >600:1, a reflectivity of less than 1% when incident normally, and a transmittance of >85%. It is connected to the output end of the lens sleeve through a first adapter ring.

[0030] In the above scheme, the stepper motor and the signal generator are connected via a BNC data cable, and the step angle is 7.5°.

[0031] In the above scheme, the stepper motor and the first adapter ring are engaged by a transmission gear to transmit the rotation angle.

[0032] A blind-structure polarization-modulated infrared thermal wave super-resolution imaging detection method includes temporal modulation of a semiconductor laser excitation source using a linear frequency modulated pulse compression signal, and spatial modulation of the laser beam using an engineering diffuser and a polarizing mirror. After the test sample absorbs the temporal-controlled excitation heat flow, the thermal wave conducts heat within the material. The change process of the surface temperature signal of the test sample is recorded by an infrared camera. A Hilbert transform is used to construct an analytical feature signal to extract the time-frequency domain feature information of the surface thermal wave signal. A feature image sequence is constructed by analyzing the instantaneous phase difference between the defective and non-defective regions. A two-dimensional Fourier transform is used to extract the high-frequency and low-frequency components of the phase detection feature image sequence under different polarization directions. The high-frequency and low-frequency components of the feature image sequence obtained from three detections are fused, and a super-resolution reconstructed image is obtained by a two-dimensional inverse Fourier transform, thereby improving the spatial resolution of the infrared thermal wave imaging detection system.

[0033] The processor sets the sampling frequency, exposure time, and acquisition time of the infrared camera; the signal generator control module drives the signal generator to output a linear frequency modulated pulse compressed signal. M ( t Connect the analog modulation signal input port of the power amplifier to enable the semiconductor laser to output pulse compression time-domain modulated laser. (1) In the formula, M ( t () is a linear frequency modulated pulse compression signal. V 0 represents the voltage of the modulation signal. f 0 is the initial frequency of linear frequency modulation. k It's frequency adjustment. t It's time.

[0034] A signal generator control module drives the signal generator to output pulse signals, which are connected to a stepper motor on the structured light modulator. At the start of each detection, the stepper motor is driven to rotate a specific angle according to a set time, which in turn drives the polarizer to rotate 120° via a transmission gear. This achieves time-domain modulation of the excitation laser pulse and spatial modulation of blind structured light in different polarization directions. Three infrared thermal imaging detections are performed under different polarization directions, and the processor controls the signal generator to output modulated signals. M ( t ), and simultaneously control the infrared camera to acquire infrared image sequences, T α ( x , y , t ) represents a specific pixel in an infrared image sequence. t Temperature value at time ( x , y ) represents the pixel position. αThe rotation angles were 0°, 120°, and 240° for the three measurements.

[0035] The thermal wave signal of the test sample surface is processed using the processor's thermal wave signal feature extraction module, and an analytical feature signal is constructed using Hilbert transform. A α ( t Its construction method is as follows: (2); In the formula, i H[.] is a complex factor, and H[.] is the Hilbert transform.

[0036] Instantaneous phase feature information of thermal wave signals is extracted using arctangent operation. The calculation method is as follows: (3); Calculate the instantaneous phase difference between the surface thermal wave signal and the linear frequency modulated pulse compression signal corresponding to each pixel. The phase difference feature is used to construct an infrared thermal wave detection image sequence; (4); The processor's super-resolution image reconstruction module processes feature image sequences under different polarization directions, and uses two-dimensional Fourier transform (FFT2) to extract the spatial spectrum of the infrared thermal wave detection phase feature image. The high-frequency components are The low-frequency component is ,in: (5); In the formula, w, v Spatial frequency of the image f max The highest spatial frequency of the feature image.

[0037] Then, the high-frequency and low-frequency components of the feature image sequence obtained from the three detections are fused at the pixel level to improve the edge features of the detection image. The fusion strategy is as follows: (6); In the formula, MAX represents the maximum value, and MIN represents the minimum value. Using... and Constructing the two-dimensional Fourier frequency of super-resolution image reconstruction Super-resolution reconstructed images are obtained through inverse two-dimensional Fourier transform (IFFT2). Achieving super-resolution reconstruction of infrared thermal wave detection images under blind structure polarization-modulated laser excitation: (7).

[0038] from Figure 3 The reconstructed image shows that the defects of the test sample are very obvious.

[0039] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A blind structure polarization-modulated infrared thermal super-resolution imaging detection method, characterized in that, A linear frequency modulated pulse compression signal is used to temporally modulate the semiconductor laser excitation source, while an engineering diffuser and polarizing mirror are used to spatially modulate the laser beam. After the test sample absorbs the temporally controlled excitation heat flow, the heat wave is conducted internally within the material. The change process of the surface temperature signal of the test sample is recorded by an infrared camera. Hilbert transform is used to construct analytical feature signals and extract the time-frequency domain feature information of the surface heat wave signal. Feature image sequences are constructed by analyzing the instantaneous phase difference between defective and non-defective regions. Two-dimensional Fourier transform is used to extract the high-frequency and low-frequency components of the phase detection feature image sequences under different polarization directions. The high-frequency and low-frequency components of the feature image sequences obtained from three detections are fused, and then a super-resolution reconstructed image is obtained by inverse two-dimensional Fourier transform.

2. The blind structure polarization-modulated infrared thermal super-resolution imaging detection method according to claim 1, characterized in that, Semiconductor lasers output pulse compression time-domain modulated lasers; (1)； In the formula, M ( t () is a linear frequency modulated pulse compression signal. V 0 represents the voltage of the modulation signal. f 0 is the initial frequency of linear frequency modulation. k It's frequency adjustment. t It is a moment.

3. The blind structure polarization-modulated infrared thermal super-resolution imaging detection method according to claim 1, characterized in that, The thermal wave signal from the surface of the test sample is processed, and an analytical characteristic signal is constructed using Hilbert transform. A α ( t Its construction method is as follows: (2)； i It is a complex factor, H[.] is the Hilbert transform, ( x , y ) represents the pixel position. α It is the rotation angle. T α ( x , y , t ) represents the pixels in an infrared image sequence. t Temperature value at any given time; Instantaneous phase feature information of thermal wave signals is extracted using arctangent operation. The calculation method is as follows: (3)； Calculate the instantaneous phase difference between the surface thermal wave signal and the linear frequency modulated pulse compression signal corresponding to each pixel. Infrared thermal wave detection image sequences are constructed using phase difference features; (4)。 4. The blind structure polarization-modulated infrared thermal super-resolution imaging detection method according to claim 1, characterized in that, Spatial spectrum of infrared thermal wave detection phase feature image extracted using two-dimensional Fourier transform The high-frequency components are The low-frequency component is ,in, (5)； In the formula, Spatial frequency of the image f max The highest spatial frequency of the feature image.

5. The blind structure polarization-modulated infrared thermal super-resolution imaging detection method according to claim 1, characterized in that, High-frequency components of the feature image sequence obtained from three detections and low frequency components Pixel-level fusion is performed using the following fusion strategy: (6)； In the formula, MAX represents taking the maximum value, and MIN represents taking the minimum value. and Constructing the two-dimensional Fourier frequency of the super-resolution reconstructed image Super-resolution reconstructed images are obtained through two-dimensional inverse Fourier transform. Achieving super-resolution reconstruction of infrared thermal wave detection images under blind structure polarization-modulated laser excitation: （7）。 6. A blind structure polarization-modulated infrared thermal super-resolution imaging detection system, applicable to the method described in any one of claims 1-5, characterized in that, The system includes an infrared camera, a processor, a signal generator, a power amplifier, a semiconductor laser, and a structured light modulator. The infrared camera is connected to the processor. One analog signal output of the signal generator is connected to the power amplifier, and the other is connected to the structured light modulator. The light spot emitted by the structured light modulator illuminates the test sample. The power amplifier is connected to the semiconductor laser, and the semiconductor laser is connected to the structured light modulator. The infrared camera, signal generator, and structured light modulator are each connected to the processor. The structured light modulator includes a lens sleeve and a collimating lens, an engineered diffuser, a polarizing mirror, and a stepper motor mounted on the sleeve. The collimating lens is fixed to the input end of the lens sleeve via a second adapter ring. The engineered diffuser is fixed inside the lens sleeve via an internal thread. The polarizing mirror is fixed to the output end of the lens sleeve via a first adapter ring. The collimating lens is connected to a semiconductor laser via an optical fiber. The stepper motor is connected to a signal generator.

7. The blind structure polarization-modulated infrared thermal super-resolution imaging detection system according to claim 6, characterized in that, The infrared camera operates at a wavelength of 3µm to 5µm, with a frame rate of ≥50Hz, an equivalent noise error of <30mK, and a field of view greater than 22°×16°.

8. The blind structure polarization-modulated infrared thermal super-resolution imaging detection system according to claim 6, characterized in that, The collimating lens is connected to the input end of the lens sleeve via a second adapter ring. The working center wavelength is 780nm, the numerical aperture NA>0.25, and the focal length f>36mm.

9. The blind structure polarization-modulated infrared thermal super-resolution imaging detection system according to claim 6, characterized in that, The engineering diffuser shapes the laser spot after it is aligned by diffuse reflection. The output spot is an approximately uniform surface heat source with a divergence angle greater than 20° and a transmittance greater than 90% in the wavelength range of 380nm~1100nm.

10. The blind structure polarization-modulated infrared thermal super-resolution imaging detection system according to claim 6, characterized in that, The polarizing mirror outputs linearly polarized light with a working wavelength of 300nm~3um, an extinction ratio of >600:1, a reflectivity of less than 1% and a transmittance of >85% when incident normally. It is connected to the output end of the lens sleeve through a first adapter ring.

Images