High-sensitivity streak camera ultrafast scanning image enhancement and noise reduction method
By introducing periodic grating modulation and lock-in amplifier demodulation into the ultrafast imaging system, the problem of noise interference in the interference fringe image was solved, the signal-to-noise ratio and contrast were improved, and the diagnostic accuracy of the imaging system was enhanced.
Patent Information
- Application Number
- CN202511669768.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-14
- Publication Date
- 2026-02-10
AI Technical Summary
In existing ultrafast imaging technologies, interference fringe images are susceptible to speckle interference, system noise, and background clutter, leading to a decrease in signal-to-noise ratio and phase distortion. The lack of front-end control means makes it difficult to effectively distinguish noise from fringe signals, and traditional post-processing methods have limited denoising effects.
By introducing a periodic grating at the imaging front end to spatially modulate the coded image, and combining the lock-in amplifier principle and ADMM-TV algorithm, physical modulation and demodulation of the image are achieved, thereby improving the contrast and signal-to-noise ratio of the interference fringe image.
It effectively separates signal and noise, significantly improves the signal-to-noise ratio and contrast of images, enhances the diagnostic accuracy and stability of imaging systems, and is suitable for ultrafast imaging applications with high spatiotemporal resolution.
Smart Images
Figure CN121509630A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application belongs to the technical field of ultrafast imaging and signal processing, and relates to a high-sensitivity streak camera ultrafast scanning image enhancement and denoising method. BACKGROUND
[0002] With the rapid development of ultrafast optical imaging technology, ultrafast imaging systems based on streak cameras have shown great application potential in the fields of physics, chemistry, biology and engineering. Such systems can record transient light field information with extremely high temporal resolution (picosecond to nanosecond level), and are particularly suitable for observing dynamic processes under extreme conditions such as inertial confinement fusion (ICF), laser plasma interaction, combustion diagnosis and light propagation in biological tissues. In the application of interferometric field imaging, for example, the CUP-VISAR system combined with the arbitrary reflecting surface velocity interferometer (VISAR) and the compressed ultrafast imaging (CUP), the core task is to accurately record the interference fringe images caused by the Doppler frequency shift effect, and to recover the velocity or phase distribution of the target from the images.
[0003] Ideally, the interference fringes should have sufficient contrast and stable phase distribution to ensure that the velocity information can be accurately recovered in the subsequent reconstruction process. However, in actual imaging process, the streak camera is inevitably affected by speckle interference caused by coherent laser illumination, detector system noise, and background clutter introduced in the encoding process, etc. Especially, the speckle interference forms high-frequency, random granular brightness fluctuations on the image, which significantly reduces the signal-to-noise ratio and phase fidelity of the interference fringe image. The existing technology relies on post-processing algorithms such as filter processing or deep learning models after image acquisition to decode and denoise the compressed observation images. These methods can improve the reconstruction quality to some extent, but still face the following key difficulties: (1) Noise and fringe signal frequency overlap, difficult to effectively distinguish. In traditional interference fringe images, the signal frequency is mainly distributed in the low frequency region, which is highly overlapped with the frequency range of typical space charge noise, making it difficult to extract the target component through frequency domain filtering, and the denoising effect is limited.
[0004] (2) Lack of front-end regulation means in the imaging process, uncontrollable noise superposition. Most current methods only perform post-processing on the acquired observation images, and cannot physically modulate the signal characteristics in the imaging stage, lacking methods to control and enhance the target signal from the source, which easily causes problems such as image contrast reduction and fringe phase distortion. SUMMARY
[0005] Therefore, the present application aims to provide a high-sensitivity streak camera ultrafast scanning image enhancement and noise reduction method, which introduces optical modulation technology to modulate the encoded image in the front-end imaging stage, improve the contrast of the interference fringe image, and provide favorable conditions for subsequent noise suppression and phase extraction.
[0006] To achieve the above object, the present application provides the following technical solutions. A high-sensitivity streak camera ultrafast scanning image enhancement and noise reduction method is applied to a streak camera-based ultrafast interference imaging system to improve the contrast and signal-to-noise ratio of the interference light field image, which comprises: spatially modulating the three-dimensional image data encoded by a spatial encoding element through a periodic grating; then compressing and imaging the modulated three-dimensional image data by a streak camera to form a two-dimensional observation image; using an ADMM-TV algorithm for image decoding to restore a three-dimensional image containing a periodic modulation structure; after reconstructing the three-dimensional image, further extracting effective fringe information in the modulation image and removing high-frequency modulation residues to realize interference fringe image denoising and enhance the imaging contrast.
[0007] Further, the encoded three-dimensional image data is denoted as Before entering the streak camera, the three-dimensional image data is modulated by a periodic grating The modulation is performed by denoted as:
[0008] In the formula, a is the modulation direct current, b is the modulation amplitude; is the grating frequency; is the initial phase, x is the horizontal coordinate, y is the vertical coordinate, t is the time; The image modulated by the periodic grating is denoted as: .
[0009] Wherein, the period of the periodic grating is set to be 1 / 8-1 / 10 of the original interference fringe period
[0010] Further, the modulated three-dimensional image data is compressed and imaged by a streak camera:
[0011] In the formula, denotes a two-dimensional observation image, denotes noise introduced by external factors, T Indicates the number of time samples of the camera within the total exposure time.
[0012] Further, the reconstructed three-dimensional image is demodulated based on a phase-locked amplifier principle to further extract effective fringe information in the modulated image and remove high-frequency modulation residues.
[0013] Further, the demodulation of the reconstructed three-dimensional image based on the phase-locked amplifier principle comprises: first, constructing two orthogonal reference signals, multiplying the reconstructed three-dimensional image respectively with the two reference signals to obtain two images; then, performing Fourier transform on the two images respectively, and filtering by using a low-pass filter; finally, synthesizing the amplitudes of the orthogonal components of the two filtered images to obtain a denoised image.
[0014] Specifically, the constructed orthogonal reference signals are respectively represented as:
[0015]
[0016] In the formula, is the initial phase, obtained according to the parameters of the periodic modulation grating; The reconstructed image is multiplied with the orthogonal reference signals respectively to obtain two images and :
[0017]
[0018] The two images and are respectively subjected to Fourier transform, and filtering by using a low-pass filter H :
[0019]
[0020] wherein, represents a two-dimensional Fourier transform, represents a two-dimensional inverse Fourier transform; through the amplitude synthesis of the orthogonal components, a denoised image is finally recovered: .
[0021] Further, the periodic grating is a detachable mask.
[0022] The present application has the beneficial effects that: the present application is aimed at the problems that the interference fringe image acquired by the fringe camera is susceptible to speckle interference, system noise and background clutter, resulting in problems such as signal-to-noise ratio reduction and phase distortion, and an ultrafast scanning image enhancement and noise reduction method based on periodic grating modulation is proposed, and the quality of the interference fringe image and the reliability of the velocity field reconstruction are effectively improved by introducing a physical modulation means in the front-end imaging stage and combining the back-end demodulation processing.
[0023] The present application realizes physical separation of the signal and the noise by moving the target signal out of the noise band through the front-end periodic grating modulation, avoiding the denoising limitations caused by frequency band overlap in the traditional frequency domain filtering. At the same time, the present application can significantly suppress high-frequency interference such as spatial charge noise while maintaining the integrity of the fringe phase structure, thereby improving the signal-to-noise ratio and contrast of the image. In addition, the method proposed in the present application can enhance the phase fidelity of the interference fringe image, providing a more reliable data basis for subsequent velocity field inversion, thereby improving the accuracy and stability of the imaging system diagnosis.
[0024] Overall, the present application provides an efficient and controllable imaging contrast enhancement method through the cooperative optimization of the imaging front-end and the processing back-end, promotes the development of high temporal and spatial resolution diagnostic technology, and is suitable for picosecond to nanosecond time scale ultrafast imaging applications based on fringe cameras, especially for diagnostic scenarios with low light field signal contrast, such as weak signal ultrafast diagnosis scenarios of biological signals or high-energy physics and astronomy signals.
[0025] Other advantages, objects, and features of the present application will be apparent to those skilled in the art from the following specification, and it is intended to be covered by the following claims. The objects and other advantages of the present application can be achieved and obtained by the following specification. BRIEF DESCRIPTION OF DRAWINGS
[0026] In order to make the objects, technical solutions and advantages of the present application clearer, the preferred embodiments of the present application will be described in detail below with reference to the accompanying drawings, in which: Figure 1 The flowchart of the high-sensitivity fringe camera ultrafast scanning image enhancement and noise reduction method provided by an embodiment of the present application is shown in the figure; Figure 2 The fringe modulation and acquisition process based on the CUP-VISAR system is shown in the figure, Figure 2 (a) is the 10th original fringe image, Figure 2 (b) is a 4x4 random binary coded aperture coding mask, Figure 2 (c) is a periodic grating pattern, Figure 2 (d) is a simulated spatial noise image,Figure 2 (e) is the noisy image after encoding and raster modulation, Figure 2 (f) is the two-dimensional observed image integrated by the fringe camera; Figure 3 is the reconstruction and demodulation result of the modulated image, Figure 3 (a) is the 10th frame image recovered by the ADMM-TV algorithm, Figure 3 (b) is the 10th frame image reconstructed based on the lock-in amplification principle; Figure 4 is the reconstruction result comparison under different intensity Perlin noise, Figure 4 (a) is the spatial structure diagram with Perlin noise intensity of 0.3, Figure 4 (b) is the reconstruction image under Figure 4 (a) condition, Figure 4 (c) is the spatial structure diagram with Perlin noise intensity of 0.4, Figure 4 (d) is Figure 4 the reconstruction image under (c) condition; Figure 5 is the reconstruction effect comparison under different fringe structure inputs, Figure 5 (a) is the 10th frame curved fringe original image, Figure 5 (b) is Figure 5 (a) the corresponding reconstruction image, Figure 5 (c) is the 10th frame sinusoidal perturbation fringe original image, Figure 5 (d) is Figure 5 (c) the corresponding reconstruction image. DETAILED DESCRIPTION
[0027] The present application is described herein with reference to specific embodiments thereof which are illustrated in the attached drawings. These are intended to demonstrate, but not limit, the application. Other advantages and novel features of the present application will become apparent from the following detailed description, given by way of example only, when considered in conjunction with the accompanying drawings. The present application can be implemented or carried out in other ways than those specifically described herein without departing from the spirit of the present application. Various modifications and variations can be made to the details of the present application without departing from the spirit of the present application, based on different viewpoints and applications. It should be noted that the drawings provided in the following examples only schematically illustrate the basic concept of the present application, and the following examples and features in the examples can be combined with each other without conflict.
[0028] The accompanying drawings are only used for illustrative purposes, and the representations are only schematic diagrams, not physical drawings, and should not be understood as limiting the present application; in order to better illustrate the embodiments of the present application, some components of the drawings may be omitted, enlarged or reduced, and do not represent the actual size of the product; it is understandable to those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.
[0029] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.
[0030] This embodiment uses the CUP-VISAR system as an example to illustrate the specific content of the method described in this invention. In the imaging path of the CUP-VISAR system, a periodic spatial grating is introduced in front of the slit of the streak camera to spatially modulate the encoded image. This grating has fixed frequency and phase parameters. The modulation process essentially introduces a reference periodic signal of fixed frequency into the spatial domain, causing the grating signal to modulate and superimpose with the interference fringe pattern. In the frequency domain, this is equivalent to a mixing effect between the signal and the reference frequency, causing the dominant frequency component of the interference fringe image to shift from the low-frequency region to near the modulation frequency. This enhances the spectral separability between the dominant frequency component of the fringe and low-frequency background interference (such as space charge noise), providing conditions for subsequent demodulation processing.
[0031] Furthermore, to effectively extract the modulated main frequency information, a frequency domain demodulation method based on the lock-in amplifier (LIA) principle is introduced. After the image is restored to a three-dimensional image sequence using a sparse reconstruction algorithm (such as ADMM-TV), an orthogonal reference signal (sine and cosine components) consistent with the grating modulation parameters is constructed and multiplied pixel-by-pixel with the reconstructed image to form a mixing signal. Subsequently, the main frequency component of the modulation band is extracted from the mixing result through Fourier transform and the application of an ideal low-pass filter. Finally, the enhanced stripe amplitude image is obtained through amplitude synthesis.
[0032] like Figure 1 As shown, this embodiment provides a method for enhancing and reducing noise in ultrafast scanning images using a high-sensitivity striated camera. The method includes: 1. Spatial modulation of the encoded 3D image data using periodic gratings.
[0033] After the probe laser illuminates the target surface, the reflected light carries phase information related to the target surface motion and forms a fringe signal in the interferometric system. This fringe light field is spatially encoded by a spatial encoding element before entering the imaging channel, and the encoded three-dimensional image data is then recorded as follows: Before entering the streak camera, it passes through a spatially periodic grating. Modulation is performed, in which is defined as: (1) wherein, a is the modulated direct current term, b is the modulation amplitude; is the grating frequency, which determines the modulation period; is the initial phase, x is the horizontal coordinate, y is the vertical coordinate.
[0034] The modulated image can be represented as: (2) 2. Compressively imaging the modulated three-dimensional image by a fringe camera.
[0035] Specifically, the modulated image sequence is integrated along the time axis in the fringe camera to form a two-dimensional observation image : (3) wherein represents the noise introduced by external factors, T represents the number of time sampling frames of the camera within the total exposure time 。
[0036] 3. Image decoding is performed by an ADMM-TV algorithm to reconstruct the image.
[0037] Since the compressive imaging process essentially belongs to an underdetermined inverse problem, in order to recover the complete three-dimensional image sequence from limited two-dimensional observation data, appropriate prior information needs to be introduced to constrain the solution space. In the reconstruction stage, the ADMM-TV algorithm is used for image decoding in this embodiment. This algorithm combines the alternating direction multiplier method with the total variation regular term, which can effectively suppress artifacts and noise in the reconstruction process while preserving fringe details, thereby recovering the three-dimensional image containing periodic modulation structure.
[0038] 4. After the three-dimensional image is reconstructed, the effective fringe information in the modulated image is further extracted and the high-frequency modulation residue is removed.
[0039] After the fringe image modulated by the grating is reconstructed, the image still contains modulation frequency components. Therefore, in order to further extract the effective fringe information in the modulated image and remove the high-frequency modulation residue, this embodiment demodulates based on the principle of a phase-locked amplifier. Specifically, the following orthogonal reference signal is constructed: (4) (5) wherein, This is the initial phase. It is obtained from the parameters of the periodic modulation grating.
[0040] Reconstruct the image Multiplying the signals by the aforementioned orthogonal reference signals yields two image streams. and : (6) (7) Two images and Perform Fourier transforms on each, and use an ideal low-pass filter. H By truncating its spectral energy, the filtered result is obtained: (8) (9) in, Represents a two-dimensional Fourier transform. This represents the two-dimensional inverse Fourier transform.
[0041] The denoised image is finally recovered by synthesizing orthogonal component amplitudes. (10) Based on the above-described stripe image denoising method, this embodiment conducted a simulation experiment in the MATLAB environment. The simulation image size was set to 512×512 pixels, and the image frame count was 20 frames.
[0042] The experimental procedure is as follows: First, a 3D image sequence containing five vertically moving stripes is generated. The spatial frequency of the stripes is approximately 0.0097, and the inter-frame movement step of the stripes along the time axis is set to 5 pixels. Figure 2 (a) shows the original striped image of frame 10 in the image sequence. A random binary coded mask with a 4×4 aperture was then introduced for modulation. Figure 2 (b) is the corresponding coded pattern; a periodic grating is superimposed at the slit position of the striped camera for spatial modulation, such as... Figure 2 (c) shows the periodic grating used, with modulation parameters of frequency 0.08 and initial phase π / 4.
[0043] To simulate space charge noise in actual observation environments, Perlin space structure noise is introduced in this embodiment, with a noise intensity of 0.2. The simulated noise is as follows: Figure 2 As shown in (d), the striped image is superimposed after being encoded with 4×4 random binary codes and modulated by a grating. Figure 2(d) The result of the shown noise is shown in Fig. 2(e), which is finally integrated in time dimension by the streak camera to obtain Figure 2 (f) The shown two-dimensional compressed observation image.
[0044] In the image reconstruction stage, the ADMM-TV algorithm is first used to reconstruct the three-dimensional image from the two-dimensional observation image. The main iteration number is set to 50, the TV denoising sub-problem iteration number is set to 20, and the regularization weight λ is set to 0.9 to retain the image edge structure and suppress the reconstruction artifacts. Figure 3 (a) shows the intermediate result of the 10th frame of the three-dimensional image recovered by the ADMM-TV algorithm, which still contains the modulated stripe structure introduced by the periodic grating. For this modulated component, a frequency domain demodulation is performed based on the principle of a lock-in amplifier. In the demodulation process, a quadrature reference signal consistent with the modulation parameters is first constructed, and an ideal low-pass filter is then applied in the frequency domain to extract the target modulation main frequency component. The low-pass filter radius is set to 6.5 pixel frequency units (i.e., the distance measured in pixels in the normalized frequency coordinate system), which corresponds to the main frequency energy distribution range near the modulation frequency. Experiments show that when the filter radius is set between 6 and 7 pixel frequency units, the modulation main frequency energy can be effectively retained, and clear stripe structure reconstruction results can be obtained. After completing the modulation main frequency component extraction and image amplitude synthesis, the original stripe image is successfully reconstructed, as shown in Figure 3 (b) shows the reconstructed 10th frame image, in which the spatial charge noise background has been effectively removed.
[0045] To further verify the adaptability and robustness of the method described in this embodiment under different experimental conditions, based on the standard simulation process shown in Figure 2 and Figure 3 , a plurality of extended experimental scenarios are constructed and tested under the premise of keeping the CUP-VISAR system structure and parameter settings unchanged. Figure 2 The typical implementation process of the method described in this embodiment under standard conditions is shown, including the generation of the original stripe image, the random binary coding modulation, the periodic grating superposition, the noise introduction, and the two-dimensional integration imaging, etc. Figure 3 The ADMM-TV algorithm reconstruction result and the demodulation processing result based on the principle of a lock-in amplifier are shown.
[0046] Based on the above standard process, only the perlin noise intensity and the stripe structure of the input image are adjusted respectively to evaluate the reconstruction performance of the system under different interference types and signal structure conditions. All tests follow a unified imaging-demodulation processing mechanism to ensure the comparability of the experimental data and the reliability of the analysis conclusion.
[0047] (1) Noise interference degree test: under the conditions of image size setting of 512*512 pixels, frame number of 20, original image of a sequence containing five vertical moving stripes, stripe spatial frequency of 0.097, inter-frame moving step of 5 pixels, random binary coding mask with aperture of 4*4 is used, and periodic grating with frequency of 0.08 and initial phase of π / 4 is used for spatial modulation. On this basis, the anti-noise performance of the system is further verified by controlling the noise intensity to simulate different complexity of observation environment. The introduced spatial structure noise is Perlin noise, and the intensity parameters are set to 0.3 and 0.4 respectively. As shown in Figure 4 Figure 4 (a) and Figure 4 (c) are two-dimensional spatial structure noise images generated when the Perlin noise intensity is 0.3 and 0.4 respectively, simulating different degrees of spatial charge interference. After the system completes the two-dimensional integral observation of the stripe image on this basis, the ADMM-TV algorithm is used for three-dimensional reconstruction, and the frequency domain demodulation is implemented based on the principle of lock-in amplifier. In the demodulation process, the orthogonal reference signal consistent with the modulation parameter is constructed, and the ideal low-pass filter is introduced in the frequency domain to extract the target modulation main frequency component. The filter radius is set to 6.5 pixel frequency units, and the center frequency position is not offset. The amplitude of the quadrature component is calculated to obtain the reconstructed stripe image, as shown in Figure 4 (b) and Figure 4 (d) are the corresponding reconstructed image results under the noise environment of Figure 4 (a) and Figure 4 (c) respectively.
[0048] From the experimental results, it can be seen that under different noise levels, the method proposed in the embodiment can still restore the main structural features of the stripe image more completely, indicating that the stripe image denoising method proposed in the application has good anti-noise performance and spectrum extraction capability, and can adapt to the imaging demand in complex spatial noise environment.
[0049] (2) Test for stripe structure change: under the conditions of image size of 512x512 pixels, frame number of 20, Perlin noise intensity of 0.2, stripe spatial frequency of 0.097, inter-frame moving step of 5 pixels, a random binary coded mask with an aperture of 4x4 is used, and a periodic grating with a superimposed frequency of 0.08 and an initial phase of π / 4 is used for spatial modulation. Under the premise of keeping all the parameters of the system consistent, two typical interference fringe patterns are selected as input images, which are curved fringes and sinusoidal perturbed fringes, respectively, to simulate different interference fringe patterns that may occur in the CUP-VISAR imaging process, and to further verify the response ability and reconstruction stability of the system to various fringe structures. FIG. 5(a) and FIG. 5(c) respectively show the input images of two typical fringe structures (both are the 10th frame). After the modulation sampling and two-dimensional integral imaging of the three-dimensional observation data are completed on this basis, the ADMM-TV algorithm is used for three-dimensional image reconstruction, and the frequency domain demodulation processing is implemented based on the principle of phase-locked amplifier. In the demodulation processing, the orthogonal reference signal consistent with the modulation parameters is constructed, and an ideal low-pass filter is introduced in the frequency domain to extract the target modulation main frequency component. The filter radius is set to 6.5 pixel frequency units, and the center frequency is not offset. Figure 5 (b) and Figure 5 (d) are Figure 5 (a) and Figure 5 (c) the 10th frame of fringe image recovered after sampling, modulation, noise superposition, three-dimensional reconstruction and frequency domain demodulation by the CUP-VISAR system.
[0050] It can be seen from the experimental results that under different fringe structure conditions, the method still can accurately extract the modulation main frequency component and effectively restore the core morphological characteristics of the interference image, which shows that the fringe image denoising method has good structural adaptability and demodulation stability for diversified interference fringes, and can meet the fringe image reconstruction requirements in complex interference scenes.
[0051] In summary, the present application provides a high-sensitivity fringe camera ultrafast scanning image enhancement and denoising method, which includes the steps of original fringe image generation, random binary coding modulation, periodic grating superposition, noise introduction and two-dimensional integral imaging. The modulation and demodulation mechanism in this method does not depend on the modification of the core optical structure of the imaging system, but only by introducing a periodic grating element (such as a detachable mask) in the imaging path and combining a frequency domain demodulation algorithm, the contrast enhancement of the interference image can be realized on the original structure of the system. The present application has good system compatibility and engineering realizability, and is suitable for picosecond to nanosecond time scale ultrafast imaging applications based on fringe camera, especially for diagnostic scenes with low contrast of optical field signals, and has high popularization and application value.
[0052] Finally, it is to be explained that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit the present application. Although the present application is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or equivalently replaced without departing from the purpose and scope of the technical solutions, and all should be covered in the scope of the claims of the present application.
Claims
1. A method for enhancing and reducing noise in ultrafast scanning images using a high-sensitivity fringe camera, applied to an ultrafast interferometric imaging system based on a fringe camera to improve the contrast and signal-to-noise ratio of the interferometric light field image, characterized in that... The method includes: spatially modulating the three-dimensional image data encoded by a spatial coding element using a periodic grating; then compressing and imaging the modulated three-dimensional image data using a fringe camera to form a two-dimensional observation image; using the ADMM-TV algorithm to decode the image and recover the three-dimensional image containing the periodic modulation structure; after reconstructing the three-dimensional image, further extracting the effective fringe information in the modulated image and removing high-frequency modulation residues to achieve noise reduction of the interference fringe image and enhance imaging contrast.
2. The method according to claim 1, characterized in that, The encoded 3D image data is denoted as Before entering the streak camera, it passes through a periodic grating. Modulation, Represented as: In the formula, a For modulating the DC term, b Modulation amplitude; The grating frequency; This is the initial phase. x The horizontal coordinate is... y For vertical coordinates, t For time; The image after periodic grating modulation is represented as follows: .
3. The method according to claim 2, characterized in that, Period of a periodic grating The period of the original interference fringes 1 / 8 to 1 / 10 of it.
4. The method according to claim 2, characterized in that, Compressed imaging of modulated 3D image data using a streak camera: In the formula, Represents a two-dimensional observation image. This represents noise introduced by external factors. T This indicates the number of time-sampled frames the camera performs during the total exposure time.
5. The method according to claim 1, characterized in that, The reconstructed 3D image is demodulated based on the lock-in amplifier principle to further extract effective stripe information from the modulated image and remove high-frequency modulation residues.
6. The method according to claim 5, characterized in that, Demodulating the reconstructed 3D image based on the lock-in amplifier principle includes: firstly, constructing two orthogonal reference signals, and then using the reconstructed 3D image... Two images are obtained by multiplying each image with a reference signal; then, Fourier transforms are performed on each image and low-pass filters are used for filtering; finally, the two filtered images are combined by orthogonal component amplitude synthesis to obtain the denoised image.
7. The method according to claim 6, characterized in that, The constructed orthogonal reference signals are represented as follows: In the formula, This is the initial phase. It is obtained from the parameters of the periodic modulation grating; Reconstruct the image Multiplying the signals by the orthogonal reference signals respectively yields two images. and : Two images and Perform Fourier transforms on each sample and apply a low-pass filter. H Perform filtering: in, Represents a two-dimensional Fourier transform. This represents the two-dimensional inverse Fourier transform; by synthesizing the amplitudes of the orthogonal components, the denoised image is finally recovered. .
8. The method according to any one of claims 1 to 7, characterized in that, The periodic grating is a detachable mask.