A terahertz imaging method, system and device based on empirical wavelet transform

Abstract

The present application relates to a terahertz imaging method, system and device based on empirical wavelet transform, in particular, the collected original terahertz signal is adaptively decomposed into the sum of several modal functions, and terahertz imaging is performed according to the maximum value of the first modal component of multiple modal functions or the amplitude value at a specific time point based on different application requirements. Compared with the existing imaging methods, the influence of system noise and environmental interference on imaging can be effectively reduced, the imaging signal-to-noise ratio can be significantly improved, and the contrast between different samples can be enhanced.

Description

Technical Field

[0001] The present invention relates to the application field of terahertz wave imaging, and in particular to a terahertz imaging method, system and device capable of reducing the influence of system noise and environmental interference. Background Art

[0002] Terahertz waves (electromagnetic waves with frequencies ranging from 0.1THz to 10THz) have low photon energy, strong penetration, and are sensitive to water absorption. The spectrum covers the vibration and rotation energy levels of a variety of organic and biological macromolecules. Terahertz imaging, especially pulse imaging technology, has broad application prospects in non-destructive testing, cultural relics protection, drug monitoring, biochemical analysis, medical diagnosis, etc. At the same time, terahertz imaging equipment is also simple, portable, and easy to integrate, so terahertz imaging has attracted more and more attention from relevant personnel. The commonly used terahertz pulse imaging methods are mainly based on the maximum value of the terahertz time domain signal obtained by scanning or the amplitude value at a specific time point, or based on the amplitude value of a certain frequency point in the spectrum after the corresponding Fourier transform of the time domain signal, or based on the absorption coefficient or refractive index of a certain frequency point obtained by comparing and calculating with the reference terahertz signal. These methods reconstruct the terahertz image of the sample.

[0003] However, due to the low intensity of existing terahertz sources, system instability, photoelectron noise, and large environmental interference, when performing terahertz imaging based on the above imaging method, the imaging quality and the contrast between different samples urgently need to be further improved. In addition, for samples with strong absorption in the terahertz band, such as biological tissue samples and optoelectronic materials, when performing terahertz imaging based on the above method, the strong absorption effect of the sample leads to a low signal-to-noise ratio of the terahertz signal, which also seriously affects the terahertz imaging quality and the distinction between different samples. Especially in the application of tomography, due to the influence of system noise and environmental interference, the weak signals of multiple reflections of layered samples will be submerged in the noise and cannot be effectively distinguished, which seriously affects the quality and resolution of terahertz tomography.

[0004] Therefore, there is an urgent need to explore new terahertz imaging methods to significantly improve the imaging signal-to-noise ratio and the contrast of different samples to meet more application requirements. Summary of the invention

[0005] Based on empirical wavelet transform, the present application adaptively decomposes the collected original terahertz signal into the sum of several modal functions, and selects the maximum value of the first modal component or the amplitude value at a specific time point for imaging according to different application requirements, thereby providing a terahertz imaging method that can effectively reduce system noise and environmental interference.

[0006] Specifically, the terahertz imaging method includes the following steps:

[0007] Step 1: Perform two-dimensional imaging on the sample to be tested to obtain the terahertz time domain signal of each point on the sample to be tested;

[0008] Step 2, performing empirical wavelet transform on the terahertz time domain signal of each point, and adaptively obtaining modal components of each resolution at each point on the measured sample;

[0009] Step 3, calculating the maximum value or the intensity at a specific time point of the first modal component among the multiple modal components;

[0010] Step 4: reconstruct the terahertz image of the sample under test.

[0011] Furthermore, step 3 also includes determining whether to perform sample surface imaging, and when performing sample surface imaging, selecting to calculate the maximum value of the first modal component, and then proceeding to step 4 to reconstruct the terahertz image. At the same time, step 3 may also include determining whether to perform sample layer tomography, and when performing sample layer tomography, determining the flight time according to the sample layer distribution, material properties and imaging depth, and selecting to calculate the intensity of the first modal component at a specific time point according to the flight time, and then proceeding to step 4 to reconstruct the terahertz image.

[0012] Preferably, the modal components of each resolution of the terahertz time domain signal of each point obtained in step 2 are: dividing the Fourier spectrum of the terahertz time domain signal into several ranges, determining the empirical scaling function and the empirical wavelet function according to the empirical wavelet, and then determining each modal component according to the above empirical scaling function and the empirical wavelet function.

[0013] The step 2 further comprises,

[0014] Step 200, calculating the Fourier spectrum of the terahertz time domain signal;

[0015] Step 201, adaptively dividing the Fourier spectrum of the signal;

[0016] Split the Fourier spectrum into N consecutive parts Λ n , define w n As the boundary of each continuous part, the midpoint of two adjacent maximum points in the Fourier spectrum of the signal is taken as the boundary:

[0017]

[0018] In order to satisfy Shannon's theorem, Λ n limit [0,π];

[0019] Step 202, determining an empirical scaling function and an empirical wavelet function;

[0020] For each wn As the center, take the width T n =2τ n As a transition segment, the empirical wavelet is defined as n The bandpass filter on the surface is constructed based on the Meyer wavelet to construct the empirical wavelet and determine the empirical scaling function and the empirical wavelet function:

[0021]

[0022]

[0023] in, and is the empirical scaling function Φ n and the empirical wavelet function The Fourier transform of n and β(x) can be expressed as τ n =rw n , 0<r<1 and β=x 4 (35-84x+70x 2 -20x 3 );

[0024] Step 203, calculate each mode function after the signal decomposition; the definition is as follows:

[0025]

[0026]

[0027] Among them, the total number of modal functions is N+1, represents the approximation coefficient, which is generated by the inner product of the signal and the empirical scaling function. is the detail coefficient of the empirical wavelet transform, which is generated by the inner product of the signal and the empirical wavelet function.

[0028] Preferably, in step 1, two-dimensional imaging is performed on the sample to be measured, using terahertz point-by-point scanning imaging or terahertz focal plane imaging, combined with transmission imaging or reflection imaging.

[0029] The present application also proposes a terahertz imaging system based on empirical wavelet transform, characterized in that the system comprises:

[0030] An imaging module performs two-dimensional imaging on the sample to obtain a terahertz time domain signal at each point on the sample;

[0031] An empirical wavelet transform module, used to calculate modal components of each resolution of the terahertz time domain signal at each point;

[0032] An image reconstruction module reconstructs the terahertz image of the sample under test according to the maximum value of the first modal component among the multiple modal components or the intensity at a specific time point.

[0033] Preferably, the empirical wavelet transform module further includes:

[0034] A Fourier spectrum calculation and segmentation module, used to segment the Fourier spectrum of a signal into N continuous parts;

[0035] An empirical scale function and empirical wavelet function determination module determines the empirical scale function and the empirical wavelet function based on the boundaries and transition segments of the N continuous parts;

[0036] And, a modal function calculation module is used to calculate and obtain each modal function after signal decomposition through the above-mentioned empirical scaling function and empirical wavelet function.

[0037] Furthermore, the system may also include a result comparison module for displaying and comparing the reconstructed terahertz images of different tested samples.

[0038] The present application also provides a terahertz imaging device, which includes a display for displaying images and processing results; a memory, and a processor connected to the display and the memory, wherein the processor is configured to: receive terahertz time domain signals of each point on the sample under test obtained by two-dimensional imaging of the sample under test; calculate modal components of each resolution of the terahertz time domain signals of each point; and reconstruct a terahertz image of the sample under test based on the maximum value of the first modal component among the multiple modal components or the intensity at a specific time point.

[0039] Compared with existing imaging methods, the terahertz imaging method based on empirical wavelet transform in this application can effectively reduce the impact of system noise and environmental interference on imaging, significantly improve the imaging signal-to-noise ratio and enhance the contrast between different samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 A flow chart of a method for terahertz imaging based on empirical wavelet transform provided in an embodiment of the present application;

[0042] Figure 2 A flow chart of performing empirical wavelet transform on terahertz time domain signals provided in an embodiment of the present application;

[0043] Figure 3 A schematic diagram of the results of performing empirical wavelet transform on the first five modal components of a terahertz time domain signal provided in an embodiment of the present application; wherein, Figure 3 a is the original signal, Figure 3 b to Figure 3 f are the first modal component signal, the second modal component signal, the third modal component signal, the fourth modal component signal and the fifth modal component signal respectively;

[0044] Figure 4 A schematic diagram of the sample surface imaging results provided in an embodiment of the present application; wherein, Figure 4 a is the original signal of a single point of the sample, Figure 4 b is the corresponding first modal component signal, Figure 4 c is the result of terahertz imaging of the fiber-reinforced polymer composite surface (uniform distribution) based on the maximum value of the reflected terahertz time-domain signal. Figure 4 d is the terahertz imaging result of the fiber-reinforced polymer composite surface (uniform distribution) based on the maximum value of the first modal component of the empirical wavelet transform.

[0045] Figure 5 A schematic diagram of the tomographic imaging results of a sample provided in an embodiment of the present application; wherein, Figure 5 a is the original signal of a single point of the sample, Figure 5 b is the corresponding first modal component signal, Figure 5 c is the result of terahertz imaging of the defective layer of the layered fiber-reinforced polymer composite material (containing defects inside) based on the amplitude value of the reflected terahertz time-domain signal at a specific time point. Figure 5 d is the terahertz imaging result of the defective layer of the layered fiber-reinforced polymer composite material (containing defects inside) based on the amplitude value of the first modal component of the empirical wavelet transform at a specific time point. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0047] The empirical wavelet transform can adaptively decompose the signal and transform it into a signal component superimposed by multiple resolutions, which can reflect the decomposed physical structure of the signal and has high time-frequency resolution and strong robustness. Based on the empirical wavelet transform, the present application adaptively decomposes the collected original terahertz signal into the sum of several modal functions. Among them, the first modal component contains the most important frequency components and effective information of the original terahertz signal, maintains a high degree of consistency and correlation with the original signal curve, and eliminates most of the influence of noise. According to different application requirements, the maximum value of the first modal component or the amplitude value at a specific time point is selected for reconstructed imaging. The following is a more detailed description of the above-mentioned and other technical features and advantages of the present application in conjunction with the accompanying drawings and embodiments.

[0048] Embodiment 1:

[0049] See the instruction manual Figure 1 As shown, it is a flow chart of the terahertz imaging method based on empirical wavelet transform, which specifically includes the following steps:

[0050] Step 1: Perform two-dimensional imaging on the sample to be tested to obtain the terahertz time domain signal of each point on the sample to be tested.

[0051] The two-dimensional imaging of the sample to be tested can be performed by terahertz point-by-point scanning imaging or terahertz focal plane imaging, combined with transmission imaging or reflection imaging. The sample here is any sample form that can be imaged by terahertz, and is not particularly limited.

[0052] Step 2: Performing an empirical wavelet transform on the terahertz time domain signal of each point to adaptively obtain the modal components of each resolution at each point on the measured sample.

[0053] See the instruction manual Figure 2 As shown, it is a flowchart for calculating the modal components of each resolution of the terahertz time domain signal at a certain point on the sample. Specifically, the modal components of each resolution of the terahertz time domain signal at each point are calculated in step 2 to obtain the Fourier spectrum of the terahertz time domain signal, divide the Fourier spectrum of the terahertz time domain signal into several ranges, determine the empirical scale function and the empirical wavelet function according to the empirical wavelet, and then determine each modal component according to the above empirical scale function and the empirical wavelet function. Specifically, it includes the following steps:

[0054] Step 200, calculating the Fourier spectrum of the terahertz time domain signal;

[0055] Step 201, adaptively dividing the Fourier spectrum of the signal;

[0056] Split the Fourier spectrum into N consecutive parts Λ n , define w nAs the boundary of each continuous part, the midpoint of two adjacent maximum points in the Fourier spectrum of the signal is taken as the boundary:

[0057]

[0058] In order to satisfy Shannon's theorem, Λ n limit [0,π];

[0059] Step 202, determining an empirical scaling function and an empirical wavelet function;

[0060] For each w n As the center, take the width T n =2τ n As a transition segment, the empirical wavelet is defined as n The bandpass filter on the surface is constructed based on the Meyer wavelet to construct the empirical wavelet and determine the empirical scaling function and the empirical wavelet function:

[0061]

[0062]

[0063] in, and is the empirical scaling function Φ n and the empirical wavelet function The Fourier transform of n and β(x) can be expressed as τ n =rw n , 0<r<1 and β=x 4 (35-84x+70x 2 -20x 3 );

[0064] Step 203, calculate each mode function after the signal decomposition; the definition is as follows:

[0065]

[0066]

[0067] Among them, the total number of modal functions is N+1, represents the approximation coefficient, which is generated by the inner product of the signal and the empirical scaling function. is the detail coefficient of the empirical wavelet transform, which is generated by the inner product of the signal and the empirical wavelet function.

[0068] Step 3: Select the first modal component and calculate the maximum value or the intensity at a specific time point;

[0069] Step 4: reconstruct the terahertz image of the sample under test.

[0070] Wherein, in step 3, different indicators are selected according to different application requirements, and then step 4 is entered to reconstruct the terahertz image. The specific selected indicators can be set according to the user's comparison requirements; and the specific selected time point can also be adjusted according to the user's comparison requirements. Specifically, when imaging the sample surface, in order to obtain the optimal image effect, the maximum value of the first modal component is selected to reconstruct the terahertz image of the sample under test; when performing sample tomography, the flight time required for the terahertz pulse to reach the imaging layer can be obtained according to the sample layered distribution, material refractive index and imaging layer depth, and the intensity of the first modal component at a specific time point is selected according to the above flight time to reconstruct the terahertz image of the sample under test. Wherein, when performing sample tomography, the flight time is calculated as follows: when imaging the interface between the N+1th layer and the Nth layer of the sample, the flight time is:

[0071]

[0072] Where c is the speed of light, n i and d i are the refractive index and thickness of the i-th layer of material respectively.

[0073] Taking the terahertz reflection imaging of fiber-reinforced polymer composite materials as an example, the imaging method of the present application is further demonstrated. Figure 3 As shown in the figure, the typical time domain signal and the result after empirical wavelet transform, the terahertz time domain signal can be adaptively decomposed into the sum of 10 modal components. For the convenience of display, Figure 3 Only the results of the first five modal components are given. Through the comparative analysis of different modal components, it can be seen that the first modal component contains the most important frequency components and effective information of the original terahertz signal, maintains a high degree of consistency and correlation with the original signal curve, and eliminates most of the noise. The calculation results of the signal-to-noise ratio and the cross-correlation coefficient are as follows: the signal-to-noise ratio of the original time domain signal is 8dB, the signal-to-noise ratio of the first modal component is 36dB, the cross-correlation coefficient between the first modal component and the original time domain signal is 0.96, and the cross-correlation coefficients between the 2nd to 5th modal components and the original time domain signal are 0.03, 0.02, 0.04, and 0.01, respectively.

[0074] In a specific embodiment, the surface (uniform distribution) of a fiber-reinforced polymer composite material is imaged by two-dimensional point-by-point scanning combined with reflection imaging, and the steps are as follows: (1) performing two-dimensional imaging of the fiber-reinforced polymer composite material to obtain the terahertz time domain signal of each point on the sample under test; (2) calculating each modal component of the time domain spectrum signal of each point on the sample under test; (3) selecting the first modal component and calculating its maximum value; (4) reconstructing the terahertz image of the surface (uniform distribution) of the fiber-reinforced polymer composite material under test based on the maximum value of the first modal component of each point on the sample.

[0075] In order to conduct comparative analysis and illustrate the advantages, the maximum amplitude value of the original terahertz time domain signal is selected at the same time to reconstruct the terahertz image of the surface (uniform distribution) of the tested fiber reinforced polymer composite material, as shown in the attached manual. Figure 4 As shown. The standard deviation and signal-to-noise ratio of the terahertz image intensity of the imaging method proposed in this application are 0.0123 and 42.3 respectively, while the standard deviation and signal-to-noise ratio of the terahertz image intensity of the traditional time domain signal maximum method are 0.0149 and 36.2 respectively. In comparison, the imaging method of this application has improved image uniformity and signal-to-noise ratio.

[0076] In another specific embodiment, a layered fiber-reinforced polymer composite material (containing defects inside) is imaged by two-dimensional point-by-point scanning combined with reflection imaging, and the steps are as follows: (1) performing two-dimensional imaging on the layered fiber-reinforced polymer composite material (containing defects inside) to obtain the terahertz time domain signal of each point on the measured sample; (2) calculating each modal component of the time domain spectrum signal of each point on the measured sample; (3) selecting the first modal component, and calculating the flight time required for the terahertz pulse to be reflected at the interface of the defect layer (corresponding to the time point 22.03 ps) based on the layered distribution of the sample, the refractive index of the material and the depth position of the defect layer in the sample, and obtaining the amplitude of the first modal component at 22.03 ps; (4) reconstructing the terahertz image of the defect layer of the layered fiber-reinforced polymer composite material (containing defects inside) under test based on the amplitude of the first modal component of each point of the sample at 22.03 ps.

[0077] In order to conduct comparative analysis to illustrate the advantages, the amplitude of the original terahertz time domain signal at 22.03ps was selected to reconstruct the terahertz image of the defect layer of the tested fiber reinforced polymer composite material (with defects inside), as shown in the attached manual. Figure 5As shown. With the original terahertz time domain signal, the defect position is not visible. With the new imaging method proposed in the present invention, the defect position is clearly visible (shown in the black dotted box), and the present invention significantly enhances the contrast between different samples. The image signal-to-noise ratio of the new imaging method proposed in the present invention at the defect is 5.9, while the image at the defect by the traditional imaging method is 2.9, and the image signal-to-noise ratio of the present invention is also greatly improved.

[0078] From the imaging results, it can be seen that since the first modal component after the empirical wavelet transform is selected in this application, when imaging is performed according to the maximum value or the amplitude value at a specific time point, compared with the traditional imaging method, the influence of system noise and environmental interference on imaging can be effectively reduced, the imaging signal-to-noise ratio can be significantly improved, and the contrast between different samples can be enhanced, achieving a beneficial technical effect. It also provides a new imaging index for terahertz imaging and promotes the development of this field.

[0079] Embodiment 2:

[0080] This embodiment provides a terahertz imaging system based on empirical wavelet transform, the system comprising:

[0081] The imaging module performs two-dimensional imaging on the sample to be tested, and obtains the terahertz time domain signal of each point on the sample to be tested. The two-dimensional imaging of the sample to be tested can be performed by using a terahertz point-by-point scanning imaging module or a terahertz focal plane imaging module, and can be combined with transmission imaging or reflection imaging. The sample here is a variety of sample forms that can be imaged by terahertz, such as biological tissue samples, composite material samples, etc., and is not particularly limited.

[0082] The empirical wavelet transform module is used to calculate the modal components of each resolution of the terahertz time domain signal at each point. Specifically, it includes obtaining the Fourier spectrum of the terahertz time domain signal, dividing the Fourier spectrum of the terahertz time domain signal into several ranges, determining the empirical scale function and the empirical wavelet function according to the empirical wavelet, and then determining each modal component according to the above empirical scale function and the empirical wavelet function. Based on the above functions, the empirical wavelet transform module can further include:

[0083] A Fourier spectrum calculation and segmentation module, used to segment the Fourier spectrum of a signal into N continuous parts;

[0084] An empirical scale function and empirical wavelet function determination module determines the empirical scale function and the empirical wavelet function based on the boundaries and transition segments of the N continuous parts;

[0085] And, a modal function calculation module is used to calculate and obtain each modal function after signal decomposition through the above-mentioned empirical scaling function and empirical wavelet function.

[0086] The image reconstruction module selects the maximum value of the first modal component or the intensity at a specific time point to reconstruct the terahertz image of the sample under test. Specifically, when performing sample surface imaging, the maximum value of the first modal component is selected to reconstruct the terahertz image of the sample under test; when performing sample tomography, the flight time is determined according to the layered distribution of the sample, material properties and imaging depth, and the intensity of the first modal component at a specific time point is selected to reconstruct the terahertz image of the sample under test. Preferably, the system also includes a result comparison module for displaying and comparing the reconstructed terahertz images of different samples under test.

[0087] In a specific embodiment, the imaging module sends the imaging signal to the empirical wavelet transform module, and the output information of the empirical wavelet transform module is sent to the image reconstruction module for reconstructing the image, and the reconstructed image information can be finally transmitted to the result comparison module for display. Further, the system also includes a control unit, for example, according to the needs of the user, the parameters selected by the image reconstruction module can be selected, or whether to display and compare through the result comparison module can be determined.

[0088] In another specific embodiment, the system may further include an input module for interacting with a user and obtaining input data from the user, such as parameter data selected by an image reconstruction module, or uploading and inputting existing imaging data, thereby realizing a flexible information input function.

[0089] In another specific embodiment, the system may further include a data output module, which can print, send or otherwise display the data results of each stage to the user for the user's use.

[0090] It should be noted that the system in this embodiment can execute the method described in Example 1.

[0091] Embodiment 3:

[0092] This embodiment provides a terahertz imaging device, which includes:

[0093] Display, used to display images and processing results;

[0094] A memory, and a processor connected to the display and the memory, wherein the processor is configured to:

[0095] The terahertz time domain signal of each point on the sample under test is received by performing two-dimensional imaging on the sample under test, and the modal components of each resolution of the terahertz time domain signal of each point are calculated. In addition, the maximum value of the first modal component or the intensity at a specific time point is selected to reconstruct the terahertz image of the sample under test.

[0096] It should be noted that the device in this embodiment can execute the method described in Example 1, or include or be equipped with the system described in Example 2.

[0097] Those skilled in the art can understand that when all or part of the above embodiments are implemented, they can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the process of the embodiments of the above methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0098] Based on empirical wavelet transform, this application adaptively decomposes the collected original terahertz signal into the sum of several modal functions. Among them, the first modal component contains the most important frequency components and effective information of the original terahertz signal, maintains a high degree of consistency and correlation with the original signal curve, and eliminates most of the noise. According to different application requirements, the maximum value of the first modal component or the amplitude value at a specific time point is selected for imaging. Compared with existing imaging methods, it can effectively reduce the impact of system noise and environmental interference on imaging, significantly improve the imaging signal-to-noise ratio and enhance the contrast between different samples.

[0099] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.

Claims

1. A terahertz imaging method based on empirical wavelet transform, characterized in that: The method comprises the following steps: Step 1: Perform two-dimensional imaging on the sample to be tested to obtain the terahertz time domain signal of each point on the sample to be tested; Step 2, performing an empirical wavelet transform on the terahertz time domain signal of each point, and adaptively obtaining the modal components of each resolution of each point on the measured sample, wherein the modal components of each resolution of the terahertz time domain signal of each point are obtained by dividing the Fourier spectrum of the terahertz time domain signal into several ranges, determining the empirical scaling function and the empirical wavelet function according to the empirical wavelet, and then determining each modal component according to the above-mentioned empirical scaling function and the empirical wavelet function; wherein step 2 further includes, Step 200, calculating the Fourier spectrum of the terahertz time domain signal; Step 201, adaptively dividing the Fourier spectrum of the signal; Split the Fourier spectrum into N consecutive parts Λ n , define w n As the boundary of each continuous part, the midpoint of two adjacent maximum points in the Fourier spectrum of the signal is taken as the boundary: In order to satisfy Shannon's theorem, Λ n Limit [0,π]; Step 202, determining an empirical scaling function and an empirical wavelet function; For each w n As the center, take the width T n =2τ n As a transition segment, the empirical wavelet is defined as n The bandpass filter on the surface is constructed based on the Meyer wavelet to construct the empirical wavelet and determine the empirical scaling function and the empirical wavelet function: Among them, and is the empirical scaling function Φ n and the empirical wavelet function 's Fourier transform, τ n and β(x) can be expressed as τ n = rw n , 0 < r < 1 and β = x 4 (35 - 84x + 70x 2 - 20x 3 ); Step 203, calculate each mode function after the signal decomposition; the definition is as follows: Among them, the total number of modal functions is N+1, It means that the approximate coefficient is generated by the inner product of the signal and the empirical scaling function. is the detail coefficient of the empirical wavelet transform, which is generated by the inner product of the signal and the empirical wavelet function; Step 3: Select the first modal component among the multiple modal components, and calculate its maximum value or intensity at a specific time point; Step 4: reconstruct the terahertz image of the sample under test.

2. The terahertz imaging method according to claim 1, characterized in that: The step 3 also includes determining whether to perform sample surface imaging, and when performing sample surface imaging, selecting to calculate the maximum value of the first modal component.

3. The terahertz imaging method according to claim 1, characterized in that: The step 3 also includes determining whether to perform sample tomography. When performing sample tomography, the flight time is determined according to the layered distribution, material properties and imaging depth of the sample, and the intensity of the first modal component at a specific time point is calculated based on the flight time.

4. The method according to any one of claims 1 to 3, characterized in that: In the step 1, two-dimensional imaging is performed on the sample to be tested, using terahertz point-by-point scanning imaging or terahertz focal plane imaging, combined with transmission imaging or reflection imaging.

5. A terahertz imaging system based on empirical wavelet transform, characterized in that: The system comprises: An imaging module performs two-dimensional imaging on the sample to obtain a terahertz time domain signal at each point on the sample; An empirical wavelet transform module is used to calculate the modal components of each resolution of the terahertz time domain signal of each point, wherein the modal components of each resolution of the terahertz time domain signal of each point are obtained by dividing the Fourier spectrum of the terahertz time domain signal into several ranges, determining the empirical scale function and the empirical wavelet function according to the empirical wavelet, and then determining each modal component according to the above-mentioned empirical scale function and the empirical wavelet function; wherein step 2 further includes, Step 200, calculating the Fourier spectrum of the terahertz time domain signal; Step 201, adaptively dividing the Fourier spectrum of the signal; Split the Fourier spectrum into N consecutive parts Λ n , define w n As the boundary of each continuous part, the midpoint of two adjacent maximum points in the Fourier spectrum of the signal is taken as the boundary: In order to satisfy Shannon's theorem, Λ n Limit [0,π]; Step 202, determining an empirical scaling function and an empirical wavelet function; For each w n As the center, take the width T n =2τ n As a transition segment, the empirical wavelet is defined as n The bandpass filter on the surface is constructed based on the Meyer wavelet to construct the empirical wavelet and determine the empirical scaling function and the empirical wavelet function: Among them, and is the Fourier transform of the empirical scale function Φ n and the empirical wavelet function , τ n and β(x) can be expressed as τ n = rw n , 0 < r < 1 and β = x 4 (35 - 84x + 70x 2 - 20x 3 ); Step 203, calculate each mode function after the signal decomposition; the definition is as follows: Among them, the total number of modal functions is N+1, represents the approximation coefficient, which is generated by the inner product of the signal and the empirical scaling function. is the detail coefficient of the empirical wavelet transform, which is generated by the inner product of the signal and the empirical wavelet function; An image reconstruction module reconstructs the terahertz image of the sample under test according to the maximum value of the first modal component among the multiple modal components or the intensity at a specific time point.

6. The system according to claim 5, characterized in that The system also includes, The result comparison module is used to display and compare the reconstructed terahertz images of different tested samples.

7. A terahertz imaging device, characterized in that: The device can execute the method as claimed in claim 1, or contain or carry the system as claimed in claim 5, and the device includes: Display, used to display images and processing results; A memory, and a processor connected to the display and the memory, wherein the processor is configured to: Receiving terahertz time domain signals of each point on the measured sample obtained by two-dimensional imaging of the measured sample; and calculating the modal components of each resolution of the terahertz time domain signal at each point; and, The terahertz image of the measured sample is reconstructed according to the maximum value of the first modal component among the multiple modal components or the intensity at a specific time point.

Images