Reflection type terahertz wave full-field imaging method based on light field homogenization

Abstract

The invention provides a reflective terahertz wave full-field imaging method based on light field homogenization, and relates to the technical field of terahertz imaging. A triangular wave is utilized to drive a galvanometer to cooperate with an f-theta lens to form a uniform illumination light field to irradiate a sample, the imaging detection module integrally moves, an intensity image is recorded, phase distribution is solved in combination with a phase recovery model based on an intensity transmission equation, and finally tilt correction is performed to obtain real complex amplitude distribution of the sample. The illumination light field can achieve full-field intensity imaging, the imaging speed is increased, meanwhile, imaging errors caused by uneven illumination are weakened, and the fidelity of imaging is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of terahertz imaging technology, and in particular to a reflective terahertz wave full-field imaging method based on optical field homogenization. Background Technology

[0002] Terahertz waves (0.1–10 THz) lie between the infrared and microwave bands, possessing advantages such as non-invasiveness, non-destructiveness, and non-ionization, making them valuable for applications in non-destructive testing, cultural relic research, and biomedicine. Compared to transmission imaging, terahertz wave reflection imaging is more suitable for characterizing the surface and near-surface structures of high-dielectric-loss materials, polar materials, and thick samples. Its imaging modes have evolved from point-by-point scanning to full-field imaging. However, existing continuous terahertz wave reflection imaging techniques often struggle to balance imaging speed and quality. Point-by-point scanning achieves high image quality and spatial resolution, but its low imaging rate makes it difficult to meet the demands of rapid detection. Full-field imaging significantly improves imaging efficiency, but it is susceptible to factors such as coherent noise, uneven illumination, and system aberrations, leading to a decline in image quality.

[0003] Furthermore, direct detection methods can only acquire intensity information, losing phase information that carries three-dimensional morphology and material characteristics. In recent years, terahertz phase reconstruction technology has made some progress, but various methods still have limitations. For example, digital holographic imaging has high requirements for light source coherence and system stability, while stacked imaging requires the acquisition of a large amount of data and has high requirements for device precision. Summary of the Invention

[0004] To address the aforementioned shortcomings in existing technologies, this invention provides a reflective terahertz wave full-field imaging method based on optical field homogenization, which balances imaging rate and imaging quality, and can achieve high-precision reconstruction of phase information of reflective samples, thereby improving the problems of decreased imaging quality and limited phase information acquisition under high-speed imaging conditions in existing technologies.

[0005] To achieve the above objectives, the technical solution adopted by this invention is as follows: a reflective terahertz wave full-field imaging method based on optical field homogenization. This reflective terahertz wave full-field imaging method is applied to a reflective terahertz wave full-field imaging system. The reflective terahertz wave full-field imaging method includes a light source module, a beam collimation module, a scanning illumination module, a sample module, and an imaging detection module. The reflective terahertz wave full-field imaging method includes the following steps: The light source module provides a stable continuous terahertz wave output; A beam collimation module is used to collimate a terahertz beam into parallel light. High-speed deflection of a terahertz beam aligned with a dual-axis scanning galvanometer is performed, and a customized... f-θThe lens performs flat-field focusing and aberration correction on the high-speed deflected terahertz beam. This involves time integration of the focused spot within one scanning cycle, and is customized accordingly. f-θ The lens's back focal plane forms a uniform illumination field whose propagation direction is approximately parallel to the optical axis; the scanning illumination module includes a dual-axis scanning galvanometer and a customized... f-θ The lens, and the dual-axis scanning galvanometers are driven by two mutually perpendicular triangular wave signals to ensure uniform scanning speed. The center of one scanning galvanometer in the dual-axis scanning galvanometer is equivalent to a customized... f-θ The center of the front focal plane of the lens; A reflective sample is used to reflect an incident uniform illumination field, causing the reflected light field to carry amplitude and phase information related to the sample surface. The sample module includes a sample stage and a reflective sample, and the sample module is placed on a customized... f-θ At the rear focal plane of the lens; After reflection, the imaging detection module generates a relative axial displacement along the optical axis of the reflective terahertz full-field imaging system, acquiring multiple full-field intensity images at different axial propagation distances. Based on the acquired full-field intensity images, the phase distribution is solved, and tilt correction is performed to obtain the true complex amplitude distribution of the sample. The imaging detection module is mounted on a single-axis adjustable displacement mechanism. The imaging detection module includes an imaging lens and a terahertz array detector, which are considered as a whole. The imaging lens images the light field carrying object information onto the terahertz array detector, and the number of acquisition frames of the terahertz array detector is set to ensure the integrity of the galvanometer scanning cycle.

[0006] Furthermore, the solution for the phase distribution includes the following steps: A focused intensity image is acquired using the imaging detection module. I And a distance from the focal position dz Defocus intensity image I dz Furthermore, by using numerical approximation to calculate the difference relationship between adjacent intensity images, the axial variation of the light field intensity is estimated. A calculation model is established based on the axial change amount to determine the relationship between phase and axial strength change amount. Based on focused intensity image I Extract the maximum light intensity of the focused intensity image. I M and with maximum light intensity I M Use the initial value to obtain the initial phase estimate. ; Based on the current phase estimate, calculate the corresponding axial strength change and compare it with the actual measured axial strength change to obtain the deviation. The deviation is used as an error correction term to update the phase distribution, and the phase error is corrected through iteration. Determine whether the corrected phase error is less than a preset threshold. J or number of iterations N When the preset upper limit is reached, the iteration is terminated, and the phase distribution at the focal plane of the imaging lens is obtained.

[0007] Furthermore, the expression for the computational model is as follows:

[0008]

[0009] in, This indicates the phase of the focusing plane of the imaging lens. k The wave vector represents the direction of propagation along the optical axis. I Indicates the focused intensity image. This represents the axial variation in light field intensity. Represents the inverse Laplace operator. This indicates the center wavelength of the terahertz wave output by the light source module.

[0010] Furthermore, the deviation The expression is as follows:

[0011] in, Represents the gradient operator. Indicates to Take the divergence.

[0012] Furthermore, the expression for the phase distribution is as follows:

[0013] in, and They represent the first n +1st time and the first n The phase distribution obtained in the next iteration Indicates the first n The deviation of the next iteration.

[0014] Furthermore, the process of obtaining the true complex amplitude distribution of the sample through tilt correction includes the following steps: The focused intensity image measured at the imaging detection module I Phase obtained from the solution The complex amplitude of the recording surface of the terahertz array detector is obtained by combining the data. O ( x , y ):

[0015] Among them, the focus intensity image I The square root term represents the amplitude of the recording plane. Represents the imaginary unit; Based on the non-uniform Fourier transform of the matrix triple product, the complex amplitude of the recording surface in the spatial domain is transformed to the frequency domain to obtain the angular spectrum distribution under tilted sampling. The expression for the spatial spectrum is as follows:

[0016]

[0017]

[0018]

[0019] in, Represents the spatial spectrum of the recording plane. and This represents the phase kernel from the spatial domain to the frequency domain. Indicates the complex amplitude distribution, ( x , y () represents the coordinates of the recording plane. K x and K y This represents the coordinates in the frequency domain of the recording area, represented by row vectors. X and Y Represents the coordinates in the spatial domain as row vectors. Represents the frequency domain coordinates of the recording plane; The corrected spectrum is obtained by performing coordinate transformation and selectively combining angular spectral components in the frequency domain. :

[0020]

[0021]

[0022] Among them, K t Represents the frequency domain coordinates after coordinate transformation. Represents the coordinate transformation matrix. and This indicates the similar angular spectrum numbers to be merged after coordinate transformation, and the corresponding angular spectrum components are... F T , m and n In the frequency domain after coordinate transformation k u , k v Number of directional similarity angular spectra Indicates the merged region; Based on the corrected spectrum The frequency domain of the object plane is transformed to the spatial domain to obtain the true complex amplitude distribution of the sample.

[0023] Furthermore, the expression for the true complex amplitude distribution of the sample is as follows:

[0024]

[0025]

[0026] in, This represents the true complex amplitude distribution of the sample. and This represents the phase kernel from the frequency domain to the spatial domain. This represents the corrected spectrum. K u and K v This represents the coordinates in the corrected frequency domain, expressed as a row vector. U and V This represents the coordinates in the corrected spatial domain, represented by row vectors.

[0027] The beneficial effects of this invention are: This invention utilizes a triangular wave driven galvanometer and f-θ A lens, in conjunction with other lenses, forms a uniform illumination field to irradiate the sample. The imaging detection module moves as a whole, recording two intensity images. The phase distribution is then solved using a phase retrieval model based on the intensity transmission equation. Finally, tilt correction is performed to obtain the true complex amplitude distribution of the sample. This invention enables full-field intensity imaging, accelerating the imaging rate while reducing imaging errors caused by uneven illumination, thus ensuring image fidelity.

[0028] The phase reconstruction method in this invention is based on the intensity transmission relationship, which describes the quantitative relationship between the phase distribution and axial intensity variation of an arbitrary monochromatic paraxial light wave field in a certain direction. Essentially, it is a quadratic elliptic partial differential equation about the phase, which can be used for partially coherent light fields. In practice, only two intensity images are needed to recover the phase. It has been applied to fields such as electricity, magnetism, and optics. Compared with other phase recovery techniques commonly used in the terahertz band, this technique does not require interference or a large amount of data redundancy, which further simplifies the optical path and operation.

[0029] The tilt correction technique involved in this invention does not use interpolation or other methods that introduce errors, but directly performs a non-uniform Fourier transform on it. This results in a more accurate correction and does not use the Jacobian determinant, which enables further large-angle tilt correction.

[0030] In summary, this invention, through the synergistic optimization of system design and algorithm innovation, breaks through the technical bottleneck of traditional continuous terahertz wave reflection imaging, providing a feasible solution for the engineering application of terahertz imaging technology, and is applicable to fields such as industrial non-destructive testing, cultural relic protection, and biomedical imaging. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the optical path of the continuous terahertz wave reflection full-field imaging system of the present invention.

[0032] Figure 2 This is a flowchart of the optical path data processing of the continuous terahertz wave reflection full-field imaging system of the present invention.

[0033] Among them, 1-light source module, 2-beam collimation module, 3-scanning illumination module, 4-sample module, and 5-imaging detection module. Detailed Implementation

[0034] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0035] Example This invention provides a reflection-based terahertz wave full-field imaging method based on optical field homogenization. This method is applied to a reflection-based terahertz wave full-field imaging system. The reflection-based terahertz wave full-field imaging method includes a light source module 1, a beam collimation module 2, a scanning illumination module 3, a sample module 4, and an imaging detection module 5. The reflection-based terahertz wave full-field imaging method includes the following steps: The light source module 1 provides a stable continuous terahertz wave output; The terahertz beam is collimated into parallel light using beam collimation module 2; High-speed deflection of a terahertz beam aligned with a dual-axis scanning galvanometer is performed, and a customized... f-θ The lens performs flat-field focusing and aberration correction on the high-speed deflected terahertz beam. This involves time integration of the focused spot within one scanning cycle, and is customized accordingly. f-θ The rear focal plane of the lens forms a uniform illumination field with its propagation direction approximately parallel to the optical axis; the scanning illumination module 3 includes a dual-axis scanning galvanometer and a customized... f-θ The lens, and the dual-axis scanning galvanometers are driven by two mutually perpendicular triangular wave signals to ensure uniform scanning speed. The center of one scanning galvanometer in the dual-axis scanning galvanometer is equivalent to a customized... f-θ The center of the front focal plane of the lens; A reflective sample is used to reflect an incident uniform illumination field, causing the reflected light field to carry amplitude and phase information related to the sample surface. Sample module 4 includes a sample stage and a reflective sample, and sample module 4 is placed in a customized... f- i At the rear focal plane of the lens; After reflection, the imaging detection module 5 generates a relative axial displacement along the optical axis of the reflective terahertz full-field imaging system, acquiring multiple full-field intensity images at different axial propagation distances. Based on the acquired full-field intensity images, the phase distribution is solved, and tilt correction is performed to obtain the true complex amplitude distribution of the sample. The imaging detection module 5 is mounted on a single-axis adjustable displacement mechanism. The imaging detection module 5 includes an imaging lens and a terahertz array detector, which are considered as a whole. The imaging lens images the light field carrying object information onto the terahertz array detector, and the number of acquisition frames of the terahertz array detector is set to ensure the integrity of the galvanometer scanning cycle.

[0036] In this embodiment, solving for the phase distribution includes the following steps: The imaging detection module 5 is used to acquire a focused intensity image. I And a distance from the focal position dz Defocus intensity image I dz Furthermore, by using numerical approximation to calculate the difference relationship between adjacent intensity images, the axial variation of the light field intensity is estimated. A calculation model is established based on the axial change amount to determine the relationship between phase and axial strength change amount. Based on focused intensity image I Extract the maximum light intensity of the focused intensity image. I M and with maximum light intensity I M Use the initial value to obtain the initial phase estimate. ; Based on the current phase estimate, calculate the corresponding axial strength change and compare it with the actual measured axial strength change to obtain the deviation. The deviation is used as an error correction term to update the phase distribution, and the phase error is corrected through iteration. Determine whether the corrected phase error is less than a preset threshold. J or number of iterations N When the preset upper limit is reached, the iteration is terminated, and the phase distribution at the focal plane of the imaging lens is obtained.

[0037] In this embodiment, tilt correction is performed to obtain the true complex amplitude distribution of the sample, including the following steps: Focused intensity images measured at 5 locations using the imaging detection module I Phase obtained from the solution The complex amplitude of the recording surface of the terahertz array detector is obtained by combining the data. O ( x , y ); Based on the non-uniform Fourier transform of the matrix triple product, the complex amplitude information is transformed to the frequency domain to obtain the angular spectrum distribution under tilted sampling; By performing coordinate transformation and selectively combining angular spectral components in the frequency domain, the corrected spectrum is obtained. ; Based on the corrected spectrum The frequency domain of the object plane is transformed to the spatial domain to obtain the true complex amplitude distribution of the sample.

[0038] In this embodiment, the core of the invention lies in constructing an imaging system with homogenized light field and optimizing phase reconstruction and tilt correction methods to achieve efficient and accurate acquisition of intensity and phase information. Its system optical path includes the following optical modules: a light source module 1 (including a continuous terahertz wave laser), a beam collimation module 2 (including a first off-axis parabolic mirror and a second off-axis parabolic mirror), and a scanning illumination module 3 (including a dual-axis scanning galvanometer and...). f-θ The system includes a lens, a sample module 4 (including a sample stage and a reflective sample), and an imaging detection module 5 (including an imaging lens and a terahertz array detector).

[0039] Light source module 1 is a continuous terahertz wave laser, which provides a stable continuous terahertz wave output, providing the energy basis for imaging; The beam collimation module 2 collimates the terahertz beam output by the continuous terahertz laser into parallel light, reducing energy loss and wavefront distortion during beam propagation. Scanning illumination module 3 includes a dual-axis scanning galvanometer and a customized... f-θ The lens, in which the dual-axis scanning galvanometer is driven by two mutually perpendicular triangular wave signals, achieves high-speed and uniform deflection of the aligned terahertz beam; customized f-θ The lens performs flat-field focusing on the deflected beam and corrects aberrations. The focused spot is integrated over time within one scanning cycle, and the focal plane then forms a uniform illumination field with the propagation direction approximately parallel to the optical axis. Sample module 4 includes a sample stage and a reflective sample, placed in a customized configuration. f-θ At the rear focal plane of the lens, the sample reflects the incident uniform illumination field, causing the reflected light field to carry amplitude and phase information related to the sample surface. The imaging detection module 5 includes a terahertz array detector and a matching imaging lens. The terahertz array detector is preferably an uncooled micro-radiometry calorimeter, but is not limited to this; other types of terahertz array detectors are also possible. The imaging lens images the light field carrying object information onto the array detector. The number of acquisition frames of the terahertz array detector is set to ensure the integrity of the galvanometer scanning cycle. The imaging lens and the array terahertz detector in this module are configured as a whole and mounted on an axially adjustable displacement mechanism. By adjusting the displacement mechanism, the imaging detection module 5 as a whole undergoes a relative axial displacement along the optical axis of the imaging system. While maintaining the spatial position of the sample under test, multiple full-field intensity images at different axial propagation distances are acquired, and the axial displacement interval does not require scaling calculation.

[0040] In this embodiment, the light source module 1 is used to provide a stable continuous terahertz wave; the output light is collimated into parallel light by the beam collimation module 2 and incident on the scanning illumination module 3; the scanning illumination module 3 includes a dual-axis scanning galvanometer and a customized... f-θ The lens, a dual-axis scanning galvanometer, consists of two dimensional galvanometers, which alter the beam's entry point. f-θ The angle of the lens is used to obtain a two-dimensional scanning light field. In order to obtain a focused beam parallel to the optical axis and a uniform illumination light field at the focal plane, the galvanometer driving signal is a mutually perpendicular triangular wave signal, and the center of the first galvanometer in the two-dimensional scanning galvanometer should be equivalently located at... f-θ The center of the front focal plane of the lens; Sample module 4 includes a sample stage and a reflective sample, centered at... f-θ On the rear focal plane of the lens, the modulated and reflected light wave field is detected by the imaging module 5. Specifically, the intensity information of the sample is imaged onto the array terahertz detector by the imaging lens.

[0041] in accordance with Figure 1 The corresponding optical system path was constructed on the cleanroom optical platform. Specifically, it consists of: a light source module 1, using a CO2-pumped terahertz laser (295 FIRL HP) with a wavelength of 118.83 μm (frequency of 2.52 THz), a maximum output power of 500 mW, and a spot diameter of 8 mm; a beam collimation module 2, composed of a pair of off-axis parabolic mirrors with a focal length of 50.8 mm; and a scanning illumination module 3, using a dual-axis gold-plated scanning galvanometer and a customized... f-θ A lens, specifically a two-dimensional gold-plated galvanometer, can linearly change the angle of the galvanometer by applying a given voltage, thereby causing the emitted beam to deflect at high speed. f-θThe lens has a front focal length of 50 mm and a rear focal length of 90 mm, and this module generates a uniform illumination field; the sample module 4 includes a sample stage and an experimental sample; the imaging detection module 5 includes an imaging lens and an uncooled micro-radiometry calorimeter. The imaging lens has a diameter of 80 mm, a focal length of 44 mm, a numerical aperture of 0.25, and the micro-radiometry calorimeter has 288 × 384 pixels, a pixel size of 35 μm × 35 μm, an effective area of ​​10.08 mm × 13.44 mm, and a refresh rate of 50 Hz.

[0042] During operation, the scanning illumination module 3 has a scanning cycle of 50 ms and an effective field of view of 2.0 cm × 2.2 cm. Beams with different deflection angles are perpendicularly incident on the focal plane to form a uniform illumination field to irradiate the experimental samples. The samples include reflective samples such as amplitude-type Siemens stars, metal coins, and printed circuit boards. The micro-amplitude calorimeter averages 5 frames of images (cumulative 100 ms) to ensure data integrity, and acquires focused intensity images and overfocus intensity images with an axial distance of 1.5 mm.

[0043] Subsequent data processing flow is as follows Figure 2 As shown, the phase recovery process is carried out according to formulas (1)-(3), and the focused intensity image and the overfocus intensity image are respectively substituted into the algorithm to obtain the phase distribution of the recording surface; the tilt correction process is carried out according to formulas (4)-(6), and finally the accurate complex amplitude distribution of the sample under uniform illumination is obtained, which will be explained in detail below.

[0044] In this embodiment, the phase reconstruction method of the present invention is based on the intensity transmission relationship. It achieves phase information inversion by analyzing the change in light field intensity along the propagation direction. This method is applicable to the partially coherent light field of the continuous terahertz wave reflection full-field imaging system of the present invention, such as... Figure 2 As shown, the specific operation is performed in conjunction with the imaging detection module: The first step is to acquire a focused intensity image (denoted as ) through the imaging detection module 5. I ) and a distance from the focal position dz The defocus intensity image (denoted as) I dz The axial variation of the light field intensity is estimated by calculating the difference between adjacent intensity images using a numerical approximation method. ; The second step is to establish a calculation model between the phase and the axial intensity change based on the relationship between the light field intensity and the propagation direction. The phase distribution can be updated through the following calculation relationship: (1) in, The wave vector representing the phase of the focusing plane of the imaging lens along the direction of optical axis propagation. , IThis indicates the focusing intensity of the imaging lens. The inverse Laplacian operator represents the axial variation of the light field intensity. The operation can be performed using a fast Fourier transform to improve computational efficiency; Step 3: Focus the intensity of the image at its maximum light intensity. I M Using the initial value, the initial estimate of the phase is obtained. ; Step 4: Calculate the corresponding axial strength change based on the current phase estimate, and compare it with the actual measured axial strength change to obtain the deviation between the two. : (2) in, Indicates the initial deviation. Represents the gradient operator. Indicates to Take the divergence.

[0045] Step 5: Introduce the deviation as an error correction term into the phase solution model to update the phase distribution, and gradually correct the phase error through iteration: (3) in, and They represent the first n +1st time and the first n The phase distribution obtained in the next iteration Indicates the first n Deviation of the next iteration n = 0, 1, 2, 3… represent the number of iterations.

[0046] In each iteration, the axial strength change is recalculated based on the updated phase estimate and compared with the actual measured value. The change is considered complete when the phase change is less than a preset threshold. J Or the number of iterations reaches the preset limit. N When the iteration ends, the accurate phase distribution at the focal plane of the imaging lens is finally obtained.

[0047] In this embodiment, in the continuous terahertz wave reflection full-field imaging system of the present invention, since the illumination beam is deflected by a two-dimensional galvanometer and obliquely incident on the surface of the reflective sample, and the imaging detection module 5 is set along a fixed direction with its imaging plane perpendicular to the system optical axis, the sample surface is not parallel to the detector image plane in the imaging coordinate system, causing geometric distortion of the imaging plane, including scale deformation and distortion. Therefore, it is necessary to perform tilt correction processing on the obtained complex amplitude after phase reconstruction to eliminate the distortion introduced by the system, thereby ensuring the accuracy of the imaging results. The present invention provides a tilt correction method for the aforementioned continuous terahertz wave reflection full-field imaging system. The process does not rely on interpolation resampling or geometric back-projection operations, but directly performs coordinate transformation and merging on the angular spectral components of the complex amplitude in the frequency domain, thereby avoiding the introduction of additional errors, such as... Figure 2 As shown, specifically, it includes the following steps: The first step is to transform the complex amplitude information obtained at the recording surface to the frequency domain based on the non-uniform Fourier transform of the matrix triple product, thus obtaining the accurate angular spectrum distribution under tilted sampling. If the coordinates of the recording surface are ( x , y The complex amplitude distribution is denoted as O ( x , y ), , If the imaginary unit is represented, then its spatial spectrum... F O It can be represented as: (4) in, and This is the phase kernel from the spatial domain to the frequency domain. K x 、K y 、X and Y These are the coordinates represented by row vectors in the frequency domain and spatial domain of the record, respectively. Represents the frequency domain coordinates of the recording plane.

[0048] The second step involves performing coordinate transformation and selectively merging angular spectral components in the frequency domain. If using... This represents the frequency domain coordinates parallel to the object plane after correction, and the corrected spectrum. F It can be represented as: (5) in, k u-m and k u-n This indicates the similar spectral numbers to be merged, and the corresponding spectral components are... F T , m、n express k u , k v Number of directional similarity angular spectra The merged region is indicated by weighted summation of the angular spectra, which compresses the data while preserving the main complex amplitude information. This step enables frequency coordinate rotation and scaling compensation, aligning the recorded surface spectrum with the object plane, while reducing frequency domain redundancy and ensuring the continuity and integrity of the complex amplitude information.

[0049] The third step involves transforming the frequency domain of the object plane to the spatial domain using the angular spectrum method and the triple product of matrices to obtain the complex amplitude distribution at the object plane, thereby recovering the true intensity and phase information of the measured sample. (6) in, and The phase kernel from the frequency domain to the spatial domain. K u 、K v 、U and V These are the coordinates represented by row vectors in the frequency domain and spatial domain, respectively. E ( u , v This is the true complex amplitude of the object plane.

[0050] In summary, the imaging uniformity, resolution, and phase fidelity of the system provided by this invention are superior to those of traditional diffractive illumination systems. Full-field intensity imaging takes only 100 ms, enabling high-speed, high-fidelity intensity and phase imaging, which meets the practical needs of industrial inspection, cultural relic protection, and other scenarios.

[0051] Although the invention has been described in detail with reference to specific embodiments, the embodiments described herein are not intended to be exhaustive or limited to the specific forms disclosed. Rather, the embodiments chosen to illustrate the problem are selected to enable those skilled in the art to practice the invention. Variations and modifications exist without departing from the spirit and scope of the invention as described and defined by the following claims.

Claims

1. A full-field imaging method for reflection-based terahertz waves based on optical field homogenization, characterized in that, The reflected terahertz wave full-field imaging method is applied to a reflected terahertz wave full-field imaging system, which includes a light source module (1), a beam collimation module (2), a scanning illumination module (3), a sample module (4), and an imaging detection module (5). The reflected terahertz wave full-field imaging method includes the following steps: A stable continuous terahertz wave output is provided using a light source module (1); The terahertz beam is collimated into parallel light using the beam collimation module (2); High-speed deflection of a terahertz beam aligned with a dual-axis scanning galvanometer is performed, and a customized... f-θ The lens performs flat-field focusing and aberration correction on the high-speed deflected terahertz beam. This involves time integration of the focused spot within one scanning cycle, and is customized accordingly. f-θ The back focal plane of the lens forms a uniform illumination field with a propagation direction approximately parallel to the optical axis; the scanning illumination module (3) includes a dual-axis scanning galvanometer and a customized... f-θ The lens, and the dual-axis scanning galvanometers are driven by two mutually perpendicular triangular wave signals to ensure uniform scanning speed. The center of one scanning galvanometer in the dual-axis scanning galvanometer is equivalent to a customized... f-θ The center of the front focal plane of the lens; The incident uniform illumination field is reflected by a reflective sample, so that the reflected light field carries amplitude and phase information related to the sample surface. The sample module (4) includes a sample stage and a reflective sample. The sample module (4) is placed in a customized environment. f- θ At the rear focal plane of the lens; After reflection, the imaging detection module (5) generates a relative axial displacement along the optical axis of the reflective terahertz full-field imaging system, and acquires multiple full-field intensity images at different axial propagation distances. Based on the acquired multiple full-field intensity images, the phase distribution is solved, and tilt correction is performed to obtain the true complex amplitude distribution of the sample. The imaging detection module (5) is mounted on a single-axis adjustable displacement mechanism. The imaging detection module (5) includes an imaging lens and a terahertz array detector, which are considered as a whole. The imaging lens is used to image the light field carrying object information onto the terahertz array detector, and the number of acquisition frames of the terahertz array detector is set to ensure the integrity of the galvanometer scanning cycle.

2. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 1, characterized in that, The solution for the phase distribution includes the following steps: A focused intensity image is acquired using the imaging detection module (5). I And a distance from the focal position dz Defocus intensity image I dz Furthermore, by using numerical approximation to calculate the difference relationship between adjacent intensity images, the axial variation of the light field intensity is estimated. A calculation model is established based on the axial change amount to determine the relationship between phase and axial strength change amount. Based on focused intensity image I Extract the maximum light intensity of the focused intensity image. I M and with maximum light intensity I M Use the initial value to obtain the initial phase estimate. ; Based on the current phase estimate, calculate the corresponding axial strength change and compare it with the actual measured axial strength change to obtain the deviation. The deviation is used as an error correction term to update the phase distribution, and the phase error is corrected through iteration. Determine whether the corrected phase error is less than a preset threshold. J or number of iterations N When the preset upper limit is reached, the iteration is terminated, and the phase distribution at the focal plane of the imaging lens is obtained.

3. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 2, characterized in that, The expression for the computational model is as follows: in, This indicates the phase of the focusing plane of the imaging lens. k The wave vector represents the direction of propagation along the optical axis. I Indicates the focused intensity image. This represents the axial variation in light field intensity. Represents the inverse Laplace operator. This indicates the center wavelength of the terahertz wave output by the light source module.

4. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 3, characterized in that, The deviation The expression is as follows: in, Represents the gradient operator, Indicates to Take the divergence.

5. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 4, characterized in that, The expression for the phase distribution is as follows: in, and They represent the first n +1st time and the first n The phase distribution obtained in the next iteration Indicates the first n The deviation of the next iteration.

6. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 1, characterized in that, The process of obtaining the true complex amplitude distribution of the sample through tilt correction includes the following steps: The focused intensity image measured at the imaging detection module (5) I Phase obtained from the solution The complex amplitude of the recording surface of the terahertz array detector is obtained by combining the data. O ( x , y ): Among them, the focus intensity image I The square root term represents the amplitude of the recording plane. Represents the imaginary unit; Based on the non-uniform Fourier transform of the matrix triple product, the complex amplitude of the recording surface in the spatial domain is transformed to the frequency domain to obtain the angular spectrum distribution under tilted sampling. The expression for the spatial spectrum is as follows: in, Represents the spatial spectrum of the recording plane. and This represents the phase kernel from the spatial domain to the frequency domain. Indicates the complex amplitude distribution, ( x , y () represents the coordinates of the recording plane. K x and K y This represents the coordinates in the frequency domain of the recording area, represented by row vectors. X and Y Represents the coordinates in the spatial domain as row vectors. Represents the frequency domain coordinates of the recording plane; The corrected spectrum is obtained by performing coordinate transformation and selectively combining angular spectral components in the frequency domain. : Among them, K t Represents the frequency domain coordinates after coordinate transformation. Represents the coordinate transformation matrix. and This indicates the similar angular spectrum numbers to be merged after coordinate transformation, and the corresponding angular spectrum components are... F T , m and n In the frequency domain after coordinate transformation k u , k v Number of directional similarity angular spectra Indicates the merged region; Based on the corrected spectrum The frequency domain of the object plane is transformed to the spatial domain to obtain the true complex amplitude distribution of the sample.

7. The full-field imaging method for reflection-based terahertz waves based on optical field homogenization according to claim 6, characterized in that, The expression for the true complex amplitude distribution of the sample is as follows: in, This represents the true complex amplitude distribution of the sample. and This represents the phase kernel from the frequency domain to the spatial domain. This represents the corrected spectrum. K u and K v This represents the coordinates in the corrected frequency domain, expressed as a row vector. U and V This represents the coordinates in the corrected spatial domain, represented by row vectors.

Images