Slit spectral imaging method based on filtering phase reconstruction and grating spectrometer

Abstract

The invention provides a slit-type spectral imaging method based on filtering phase reconstruction and a grating spectrometer. The method comprises the following steps: measuring wavefront aberration of a slit-type grating imaging spectrometer before slit filtering by using a wavefront sensor; based on a slit filtering principle, reconstructing a phase function after slit filtering by using wavefront aberration before slit filtering; and capturing a slit image, and reconstructing the slit image by using the reconstructed filtered phase function of the slit to obtain a final reconstructed image. According to the method, effective recovery of the slit imaging data of the grating spectrometer based on the slit is realized, the influence of optical aberration and image noise on the spectrum and spatial resolution of the spectrometer is effectively eliminated, and the slit imaging quality is remarkably improved and is close to the diffraction limit.

Description

Technical Field

[0001] The present invention belongs to the field of spectral analysis technology and is used for a method and system for real-time monitoring of ink residue on a grinding device, in particular to a slit-type spectral imaging method and a grating spectrometer based on filtered phase reconstruction. Background Art

[0002] The constantly evolving solar magnetic field plays a dominant role in all processes in the solar atmosphere. It generates most of the spectacular visible phenomena, such as sunspots, prominences, solar flares, and coronal mass ejections. The solar atmosphere can contribute to severe space weather. To predict and forecast space weather, it is crucial to develop an accurate physical model of the solar atmosphere and to monitor and forecast solar activity events in real time (e.g., optical thickness, velocity, oscillation frequency, etc.). High-spectral-resolution grating spectrometers are crucial for studying the thermodynamic properties of solar activity.

[0003] Grating spectrometers are among the most important instruments for solar observations. Reliable spectroscopic measurements of the Sun's fine structure require grating spectrometers with high spatial, spectral, and temporal resolutions. Typically, accurate spectral information can be obtained through spectral calibration, while large-field-of-view images of slit-based spectrometers can be obtained by stitching slit images. However, the quality of the captured images is significantly degraded due to imaging noise and optical aberrations.

[0004] Adaptive optics (AO) can compensate for optical aberrations caused by atmospheric turbulence in real time. Therefore, the quality of the slit image depends largely on the correction performance of the adaptive optics system. Unlike general optical systems, the slit of a grating spectrometer is narrow enough, so the slit will have a certain degree of filtering effect on the wavefront aberration. Traditional adaptive optics methods cannot be directly applied to correct the optical aberrations of a grating spectrometer. In order to solve the above problems, an aberration correction method for a grating imaging spectrometer based on adaptive optics is proposed. In order to accurately detect static aberrations, a calibration system is used. In addition, a differential adaptive optics correction method based on the sensitivity of the filter slit is proposed. However, traditional adaptive optics is not perfect because it is limited by the isovain zone and the complexity of adaptive optics.

[0005] In addition to adaptive optics correction methods, image reconstruction techniques can also eliminate the effects of atmospheric turbulence and obtain diffraction-limited images over a large field of view. The effectiveness and accuracy of image restoration algorithms are reduced due to the presence of optical aberrations and noise. Optical aberrations can introduce non-uniform and spatially varying effects in blurred images, leading to point spread function (PSF) and camera detection errors, which can have a particularly adverse impact on deep learning-based image classification results. Furthermore, noise introduces random variations in pixel intensity, reducing contrast and potentially producing false structures or artifacts. This makes it difficult for restoration algorithms to accurately distinguish between true image features and noise-induced variations. Several deblurring methods based on image priors and PSF priors have been proposed, but unfortunately, the estimated PSF still contains errors and is prone to artifacts. Furthermore, most algorithms consider PSF errors, but rarely discuss the impact of target detail loss on the algorithm, such as the loss of certain structure and phase shifts in the slit image caused by slit filtering.

[0006] Unlike traditional optical systems, the filtering slits cause some structure in the slit image to be lost and the phase to shift. Existing technologies for these two aforementioned categories are primarily categorized into two types: adaptive optics and post-processing. Traditional adaptive optics cannot be directly applied to correct aberrations and noise in slit-grating spectrometers. Furthermore, traditional adaptive optics is imperfect due to limitations in isovain zones and the complexity of adaptive optics. Furthermore, commonly used deblurring methods are inefficient or even ineffective in slit image reconstruction.

[0007] Therefore, how to provide a slit grating spectrometer and a spectral imaging method that can accurately reconstruct the phase when used for slit image restoration is an urgent problem that needs to be solved by those skilled in the art. Summary of the Invention

[0008] In view of this, the present invention proposes a slit-type spectral imaging method and a grating spectrometer based on filtered phase reconstruction, which are used to solve the problem that the quality and performance of spectral reconstruction are limited by the influence of aberrations and noise, and realize the effective recovery of slit imaging data of a slit-based grating spectrometer, effectively eliminating the influence of optical aberrations and image noise on the spectral and spatial resolution of the spectrometer, significantly improving the slit imaging quality and approaching the diffraction limit.

[0009] In order to achieve the above object, the present invention adopts the following technical solutions:

[0010] The present invention first discloses a slit-type spectral imaging method and a grating spectrometer based on filtered phase reconstruction, comprising the following steps:

[0011] S1: Using a wavefront sensor to measure the wavefront aberration before slit filtering in a slit grating imaging spectrometer;

[0012] S2: Based on the slit filtering principle, the wavefront aberration before slit filtering is used to reconstruct the phase function after slit filtering;

[0013] S3: Capturing a slit image, and reconstructing the slit image using the reconstructed phase function after slit filtering to obtain a final reconstructed image.

[0014] Preferably, the S1 comprises the following steps:

[0015] S11: Split the incident beam into multiple small wavefronts using a microlens array;

[0016] S12: measuring the focus offset before extracting each wavelet;

[0017] S13: Calculating the wavefront slope according to the center coordinates of each microlens and the focus offset of its corresponding wavelet;

[0018] S14: The wavefront phase function is converted into a Zernike polynomial expansion form using the wavefront slope, and the polynomial coefficients are fitted using the least squares method. The wavefront aberration before slit filtering is quantified based on the polynomial coefficients.

[0019] Preferably, the Zernike polynomial expansion form is:

[0020]

[0021] Where φ(x,y) is the wavefront phase function, a n is the coefficient of the polynomial to be determined, Z n (x,y) is the nth order Zernike polynomial, (x,y) is the focal coordinate of the wavelet front;

[0022] Use the least squares method to fit the polynomial coefficient a to be determined n , such that:

[0023]

[0024] Where s x (x i ,y i ),s y (x i ,y i ) is the coordinate of the i-th wavelet front intersection point (x i ,y i ) corresponds to the x-axis wavefront slope and y-axis wavefront slope.

[0025] Preferably, S2 comprises the following steps:

[0026] S21: Calculate the near-field function E of the optical path before slit filtering based on the wavefront aberration before slit filteringnf (x,y);

[0027] S22: through the near-field function E nf The far-field function E of the optical path before slit filtering is obtained by Fourier transform of (x,y) far (x,y);

[0028] S23: Through the far-field function E far The Fourier transform of (x,y) reconstructs the phase function after slit filtering.

[0029] Preferably, the phase function after slit filtering reconstructed in S23 is Expressed as:

[0030]

[0031] Where, the functions real{} and angle{} return the real part and phase angle of the complex array respectively, F{} represents the Fourier transform, pupil(x,y) represents the pupil aperture function, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, i is the imaginary unit, w s is the width of the slit, and Ls is the length of the slit.

[0032] Preferably, S3 includes the following steps:

[0033] S31: Capture the slit image and use it as the blurred image O(x,y);

[0034] S32: Reconstruct the point spread function (PSF) based on the phase function after slit filtering;

[0035] S33: Iteratively update the estimated value I of the blurred image O(x,y) using the reconstructed point spread function PSF n+1 (x, y), i.e., the final reconstructed image;

[0036] Preferably, the S33 uses the LRRA algorithm to perform image reconstruction, which is expressed as:

[0037]

[0038] Where A0 is the initial value or reference amplitude of the wavefront amplitude, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, and ⊙ is the nonlinear reconstruction operator, defined as:

[0039]

[0040] in, is the Fourier transform of A, F -1is the inverse Fourier transform, and α and β are adjusted between -1 and 1 with the goal of minimizing noise.

[0041] The present invention also provides a slit grating spectrometer according to the slit grating spectral imaging method based on filtered phase reconstruction, comprising: a laser light source module and a beam splitter arranged in sequence along the propagation direction of the light path, the beam splitter divides the incident light into two paths, the first light path passes through an aberration generator to generate a preset phase distribution and is reflected and fused to the second light path, the wavefront sensor measures the wavefront aberration before the slit filtering after being modulated by the aberration generator in the second light path; after passing through the slit, the second light path is collimated by an off-axis parabolic mirror and projected onto a reflective grating; the reflective grating disperses light in different directions according to different wavelengths to form a spectral image; the detector captures the spectral image and sends it to the image processing and algorithm module to reconstruct the slit image to obtain the final reconstructed image.

[0042] Preferably, the laser light source module adopts a monochromatic laser.

[0043] Preferably, the following are arranged in sequence between the laser light source module and the beam splitter along the light propagation direction:

[0044] Pinhole, used to filter out stray light in the outgoing beam of the laser light source module;

[0045] The collimator is used to collimate the light beam after passing through the pinhole, forming parallel light and then incident on the beam splitter.

[0046] It can be seen from the above technical solution that, compared with the prior art, the beneficial effects of the present invention include:

[0047] Improved spectral and spatial resolution: Through precise measurement and correction of wavefront aberrations and analysis of point spread functions (PSFs) under different slit width conditions, the present invention can achieve high spectral and spatial resolution in slit imaging, which is crucial for scientific research and technological development.

[0048] Enhanced image reconstruction quality: The LRRA algorithm effectively suppresses noise and restores image blur caused by aberrations, ensuring high-quality spectral images. Experimental results show that this method achieves good reconstruction results under different aberration levels, especially in the case of large aberrations.

[0049] Effectiveness over a wide spectral range: The present invention is not only applicable to monochromatic light sources, but also works effectively over a wide spectral range and can accurately restore slit images even under white light conditions, which makes it highly adaptable and practical in a variety of application scenarios.

[0050] Promoted the development of high spectral resolution grating spectrometer technology: By bringing the imaging performance of the optical system close to the diffraction limit, the present invention provides strong support for improving the overall performance of the spectrometer, which is of great significance for promoting the development of high spectral resolution grating spectrometer technology.

[0051] In summary, the present invention applies the LRRA algorithm, combined with the phase reconstruction method and Hartmann wavefront detection technology, to achieve effective recovery of slit imaging data of a slit-based grating spectrometer, effectively eliminating the influence of optical aberrations and image noise on the spectral and spatial resolution of the spectrometer, significantly improving the slit imaging quality and approaching the diffraction limit.

[0052] This method is applicable not only to scenarios with large optical aberrations and low signal-to-noise ratios, but also plays an important role in wide-spectrum and even white-light scenarios. The proposed method accurately reconstructs slit images, bringing the imaging performance of optical systems close to the diffraction limit. This has important implications for the development of high-quality slit image stitching and grating-based high-spectral-resolution spectrometers.

[0053] By solving the above-mentioned technical problems, the invention not only improves the performance of the imaging grating spectrometer, but also broadens its application field, showing obvious superiority. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only embodiments of the present invention. Those skilled in the art can also derive other drawings based on the provided drawings without inventive effort.

[0055] Figure 1 A flow chart of a slit grating spectral imaging method based on filtered phase reconstruction provided by an embodiment of the present invention;

[0056] Figure 2 A light path diagram of a grating spectrometer system provided in an embodiment of the present invention;

[0057] Figure 3 The reconstructed Zernike polynomial coefficients of different orders provided by the embodiment of the present invention ( and ) Schematic diagram;

[0058] Figure 4 Schematic diagram of the distribution of unreconstructed wavefront aberration and wavefront aberration reconstructed through a filtering slit provided in an embodiment of the present invention;

[0059] Figure 5Schematic diagram of the effect of different slit widths on the PSF two-dimensional distribution image provided by an embodiment of the present invention;

[0060] Figure 6 The embodiment of the present invention provides Schematic diagram of the result of image restoration using the LRRA algorithm in this case;

[0061] Figure 7 The embodiment of the present invention provides Schematic diagram of the result of image restoration using the LRRA algorithm in this case;

[0062] Figure 8 A comparison chart of white light spectrum image restoration results provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0064] Thermal imaging technology, as a non-contact, highly sensitive method for visualizing temperature fields, has been successfully applied in equipment condition monitoring across multiple industrial sectors. It can display the temperature distribution on an object's surface in real time by detecting the infrared radiation emitted by the object itself.

[0065] Robust infrared image deblurring methods, such as Lucy-Richardson (LR) deconvolution, ForWaRD, LRRA, and deep learning methods, are used to restore blurred images. Among these deblurring methods, the Lucy-Richardson-Rosen Algorithm (LRRA) is a novel computational reconstruction method developed by combining the well-known Lucy-Richardson algorithm (LRA) and nonlinear reconstruction (NLR). Many studies have shown that LRRA significantly outperforms LRA and NLR. These studies mainly focus on how to strike an appropriate trade-off between improving image clarity and suppressing noise. However, most algorithms consider the error in the PSF, but rarely discuss the impact of target detail loss on the algorithm.

[0066] like Figure 1 As shown, the slit grating spectral imaging method based on filtered phase reconstruction provided by the first aspect of the embodiment of the present invention mainly includes the following steps:

[0067] S1: Using a wavefront sensor to measure the wavefront aberration before slit filtering in a slit grating imaging spectrometer;

[0068] S2: Based on the slit filtering principle, the wavefront aberration before slit filtering is used to reconstruct the phase function after slit filtering;

[0069] S3: Capture the slit image and reconstruct the slit image using the reconstructed slit filtered phase function to obtain the final reconstructed image.

[0070] In one embodiment, S1 includes the following steps:

[0071] S11 Incident wavefront segmentation: A microlens array is used to split the incident beam into multiple small areas. Each microlens corresponds to a sub-aperture and focuses its corresponding local wavefront.

[0072] S12 Focus Imaging and Image Acquisition: Each microlens focuses the light beam onto an image sensor (such as a CCD or CMOS). If the wavefront is an ideal plane wave, each microlens forms a regularly arranged light spot on the focal plane. If the wavefront is distorted, the positions of these light spots will deviate from their "reference" positions. The image sensor records the actual coordinates of all focal points and extracts the position offset of each focal point using image recognition algorithms (such as centroid detection and Gaussian fitting).

[0073] S13 calculates the local wavefront slope:

[0074] Assume that the center coordinates of each microlens are (x i ,y i ), in the absence of distortion, the focus should be at the ideal position (x i0 ,y i0 ).

[0075] The actual detected focus position is (x if ,y if ), the offset is:

[0076] Δx i =x if -x i0 ;

[0077] Δy i =y if -y i0 ;

[0078] According to the geometric optics relationship, the slope of the local wavefront at this sub-aperture is:

[0079] s x (x i ,y i )=Δx i / f;

[0080] s y (xi ,y i )=Δy i / f;

[0081] Where f is the focal length of the microlens.

[0082] S14: Wavefront phase distribution reconstruction:

[0083] Known wavefront slope distribution s x (x,y) and s y (x, y), the wavefront phase function φ(x, y) can be recovered by integration. This is a typical gradient field integration problem. The specific steps are as follows:

[0084] First, the objective function is solved using the least squares reconstruction method:

[0085] Objective function:

[0086]

[0087] The solution method is: construct a system of linear equations and use matrix operations or iterative methods to solve them.

[0088] Secondly, the Zernike polynomial is fitted and the wavefront is expressed as a Zernike polynomial expansion form:

[0089]

[0090] Where Z n (x,y) is the nth order Zernike polynomial, a n is the coefficient to be determined.

[0091] Finally, the least squares method is used to fit the coefficient a n , such that:

[0092]

[0093] Fitting coefficient a n It can accurately describe the aberration of the wavefront and can be used to analyze the type of wavefront aberration (such as spherical aberration, coma, astigmatism, etc.).

[0094] In one embodiment, the Zernike polynomial expansion is:

[0095]

[0096] Where φ(x,y) is the wavefront phase function, a n is the coefficient of the polynomial to be determined, Z n(x,y) is the nth order Zernike polynomial, (x,y) is the focal coordinate of the wavelet front;

[0097] Use the least squares method to fit the polynomial coefficient a to be determined n , such that:

[0098]

[0099] Where s x (x i ,y i ),s y (x i ,y i ) is the coordinate of the i-th wavelet front intersection point (x i ,y i ) corresponds to the x-axis wavefront slope and y-axis wavefront slope.

[0100] In one embodiment, S2 applies filtering technology to reconstruct the calculated phase to remove noise and unnecessary interference and improve the accuracy of the phase information. The steps include:

[0101] S21: Calculate the near-field function E of the optical path before slit filtering based on the wavefront aberration before slit filtering nf (x,y);

[0102] S22: through the near-field function E nf The far-field function E of the optical path before slit filtering is obtained by Fourier transform of (x,y) far (x,y);

[0103] S23: Through the far-field function E far The Fourier transform of (x,y) reconstructs the phase function after slit filtering.

[0104] In this embodiment, it is assumed that the aberrations before and after slit filtering are and Then the near-field function and far-field function of the optical system can be expressed by E nf (x,y) and E far (x,y) represents the near field function Enf(x,y) can be expressed as:

[0105]

[0106] Where pupil(x,y) represents a pupil aperture function, which is expressed as:

[0107]

[0108] Where r represents the radius of the aperture.

[0109] Far field function Efar (x,y) can be expressed as:

[0110] E far (x,y)=F{E nf (x,y)}×rect(x / w s )×rect(y / L s );

[0111] In the formula, the function rect(x / w s ) and rect(y / L s ) can be expressed as:

[0112]

[0113] Due to the slit filtering effect, part of the wavefront aberration will be filtered out, so it is necessary to reconstruct the filtered phase. nf-s (x,y) can be calculated using the following formula:

[0114] E nf-s (x,y)=F{E far (x,y)};

[0115] Wherein, the subscript s represents the slit.

[0116] The filtered phase can be expressed as:

[0117]

[0118] Where, the functions real{} and angle{} return the real part and phase angle of the complex array respectively, F{} represents the Fourier transform, pupil(x,y) represents the pupil aperture function, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, i is the imaginary unit, w s is the width of the slit, and Ls is the length of the slit.

[0119] In one embodiment, S3 combines the Lucy-Richardson-Rosen algorithm to restore the slit image. This algorithm can effectively suppress noise and improve image clarity, and is particularly suitable for situations with a large number of optical aberrations and low signal-to-noise ratio. It includes the following steps:

[0120] S31: Use a high-resolution camera to capture the slit image, normalize the acquired slit image, usually scaling its value range to [0,1], and use it as the blurred image O(x,y); the final image reconstruction process can be expressed as:

[0121]

[0122] Where O(x,y) is the blurred image, I(x,y) is the clear image, PSF(x,y) is the point spread function, and n(x,y) is the noise.

[0123] S32: Reconstruct the point spread function (PSF) based on the phase function after slit filtering;

[0124] S33: Iteratively update the estimated value I of the blurred image O(x,y) using the reconstructed point spread function PSF n+1 (x,y), the final reconstructed image;

[0125] In one embodiment, S33 uses the LRRA algorithm to reconstruct the image, and through multiple iterations of optimization, gradually improves the image quality until the desired resolution and clarity are achieved. The reconstructed image of the (n+1)th iteration can be expressed as:

[0126]

[0127] Where A0 is the initial value or reference amplitude of the wavefront amplitude, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, and ⊙ is the nonlinear reconstruction operator, defined as:

[0128]

[0129] in, is the Fourier transform of A, F -1 For the inverse Fourier transform, α and β are adjusted between -1 and 1 to minimize the noise.

[0130] In the first iteration, set I1 = O(x, y), normalize it, and then calculate O(x, y) and The ratio between In the next step, this “ratio” is compared with the PSF using the NLR (non-linear correlation) method. s (x, y) is subjected to nonlinear correlation processing, resulting in a result called the "residue." This "residue" is multiplied by the result of the previous iteration, i.e., I after the nth iteration. This process is iterated until the optimal reconstruction result is achieved. Because NLR provides a better estimate, the algorithm converges quickly. In subsequent verification of the algorithm, the values ​​of α and β were set to 0.9 and 1.0, respectively. This example does not discuss the impact of the selection and optimization of these parameters on algorithm performance.

[0131] The embodiment of the present invention is not limited to the LRRA algorithm, and may also adopt LRA (Lucy-Richardson Algorithm) and an image restoration method based on regularization technology.

[0132] The present invention also provides a slit grating spectrometer based on the slit grating spectral imaging method based on filtered phase reconstruction, comprising: a laser light source module and a beam splitter arranged in sequence along the propagation direction of the light path, the beam splitter divides the incident light into two paths, the first light path passes through an aberration generator to generate a preset phase distribution and is reflected and fused to the second light path, the wavefront sensor measures the wavefront aberration before the slit filtering after being modulated by the aberration generator in the second light path; after passing through the slit, the second light path is collimated by an off-axis parabolic mirror and projected onto a reflective grating; the reflective grating disperses light in different directions according to different wavelengths to form a spectral image; the detector captures the spectral image and sends it to the image processing and algorithm module to reconstruct the slit image to obtain the final reconstructed image.

[0133] In one embodiment, the laser light source module uses a monochromatic laser.

[0134] In one embodiment, a Hartmann-Shack wavefront sensor (HSWFS) is used as the wavefront sensor. The HSWFS is first used to measure wavefront aberrations in a grating imaging spectrometer system. This sensor reconstructs the shape of the entire wavefront by splitting the incident light beam into multiple wavefronts and measuring the offset of each wavefront.

[0135] In one embodiment, Figure 2 As shown, the optical path structure is:

[0136] Laser 1 provides a stable monochromatic or broadband light source. Examples include polarized HeNe lasers (wavelength 632.8 nm, full width at half maximum (FWHM) 1 pm) and fianium white light lasers. The operating principle is to illuminate a sample or test subject with a light source of a specific wavelength for spectral analysis. For example, in experiments, a polarized HeNe laser was used to simulate a stable light source, and in some experiments, a fianium white light laser was used to verify the effectiveness of the method.

[0137] Pinhole 2 is used to filter out stray light and improve beam quality.

[0138] Collimator,3,collimates the light beam after passing through the pinhole to form parallel light.

[0139] The non-polarizing beam splitter 4 is used to split the incident light into two parts. One part passes through the LC-SLM 5 to produce a preset phase distribution, and the other part goes directly to the Hartmann wavefront detector (HS WFS) 6 for detection. This allows researchers to generate specific aberration patterns and accurately measure and correct them.

[0140] The reflective liquid crystal (LC-SLM) 5 is used to perform phase modulation on the collimated light beam and adjust the wavefront shape as needed.

[0141] The Hartmann wavefront detector 6 uses a microlens array to sample the incident light field, calculate the tilt angle within each subaperture, and reconstruct the shape of the entire wavefront. By measuring the wavefront information after LC-SLM modulation, it provides data support for subsequent aberration correction. By detecting the controllable aberrations generated by the LC-SLM and the HS WFS, effective correction of optical system aberrations is achieved, improving imaging quality.

[0142] The imaging mirror 7 and the slit 8 are used to further process the light beam and may be used for specific spectral analysis or imaging tasks. The width of the slit 8 is adjustable, and the slit design enhances the resolution within a specific wavelength range, making it suitable for fine spectral analysis tasks.

[0143] Off-axis parabolic mirror 9 and grating 10: Incident light first passes through slit 8 to limit its size, then is collimated by off-axis parabolic mirror 9 and projected onto grating 10. The grating disperses the light in different directions according to its wavelength, forming a spectral image and achieving spectral decomposition. The width of the slit affects the quality of the final image and the degree of detail retained.

[0144] Reflector 11 and detector 12, which includes a CCD / CMOS camera and a bandpass filter. The operating principle is that light reflected by reflector 11 is focused on detector 12, where it is captured and digitized to produce the final spectral data. The bandpass filter is used to achieve order sorting, ensuring accurate spectral information.

[0145] The image processing and algorithm module further processes the acquired data to restore high-quality images. The LRRA algorithm corrects image degradation caused by optical system limitations or other factors, improving image clarity and detail. In particular, after accounting for the slit filter effect, accurate reconstruction of the filter phase is essential for effective image reconstruction.

[0146] The present invention uses spectral reconstruction technology to decompose the grating spectrum into a two-dimensional slit image and one-dimensional spectral information. By verifying the reconstructed image, it is ensured that its resolution in both spatial and spectral dimensions has been significantly improved.

[0147] HeNe lasers are used as light sources. Due to their narrow bandwidth, measured in the picometer range, they can be considered monochromatic in almost all spectral systems. Therefore, spatial resolution can be verified by performing image restoration on monochromatic light. Furthermore, using a white light laser as the light source, spectral resolution can be verified. Finally, the image restoration results are compared with the adaptive optics correction results to verify the effectiveness of the present invention.

[0148] Figure 3 The horizontal axis Zernike order of the histogram is the order of the Zernike polynomial, and the vertical axis Coefficients is the coefficient of the Zernike polynomial. Figures (a) and (b) show two different orders (Figure (a) is Figure (b) is ), which are used to describe various aberration types in optical systems, such as defocus, coma, and spherical aberration. The height of each column represents the contribution of the corresponding order of Zernike mode to the total wavefront aberration.

[0149] Figure 4 The significant differences between the unreconstructed image, the reconstructed image, and the experimentally obtained image are shown; slits of different widths are used. Figures (a) and (b) represent and Under the condition of 0dA slit wavefront phase distribution; Figure (c) and Figure (d) respectively represent and Under the condition of 1dA slit wavefront phase distribution; Figure (e) and Figure (f) respectively show and In this case, the 2dA slit wavefront phase distribution. Among them, the marks "0dA", "1dA" and "2dA" respectively represent the absence of filter slit, the slit width is 1 Airy disk diameter (w s =1dA) and the slit width is 2 Airy disk diameters (w s =2dA). "dA" represents the diameter of the Airy disk. PV represents the difference between the maximum and minimum phase deviations of the wavefront. RMS represents the root mean square value of the wavefront phase deviation and is often used to measure the overall level of aberration. As can be seen from the figure, the reconstructed image is highly consistent with the experimental image, indicating that the unreconstructed filtered phase results in a mismatch between the slit image and its corresponding phase, rendering commonly used deblurring techniques inefficient or even ineffective for slit image reconstruction.

[0150] Figure 5 The image shows the two-dimensional distribution of the point spread function (PSF) in an optical system, which is usually used to evaluate the performance and aberration of the imaging system. Specifically, it contains multiple sub-graphs, each of which represents the PSF distribution under different conditions or after different processing steps. The first row in the figure shows the distribution of the PSF under small aberrations. In this case, the slit width is one Airy disk diameter (w s =1dA) in different processing stages or different correction methods. The second row in the figure shows the small aberration In this case, the slit width is 2 Airy disk diameters (w s =2dA) in different processing stages or different correction methods. The third row in the figure shows the large aberration In this case, the slit width is one Airy disk diameter (w s =1dA) in different processing stages or different correction methods, the fourth row in the figure shows the large aberration In this case, the slit width is 2 Airy disk diameters (w s =2dA) in different processing stages or different correction methods.

[0151] Among them, the left column shows the original or uncorrected PSF distribution. It can be seen that the light spot is relatively blurred, indicating that there is a certain degree of aberration. The middle column shows the PSF distribution after some preliminary correction or processing. Compared with the left side, the light spot has become more concentrated, indicating that the aberration has been improved to a certain extent. The right column shows the final correction result or the PSF distribution under the ideal state. At this time, the light spot is the clearest and most concentrated, close to the imaging effect of an ideal point light source. The color bar on the right side of each sub-image represents the intensity distribution, and the change from blue to red represents the change from low to high intensity. Through the change of color, you can intuitively see the distribution of light intensity in space. Each sub-image uses a pair of red dotted lines to represent the position of the slit image, and the coordinate axis unit is pixels.

[0152] Figure 6 Shows small aberrations Under the same conditions, the performance indicators of the slit images reconstructed using the LRRA algorithm at different slit widths, such as SSIM, MSE, ISNR and PSNR, prove the effectiveness and stability of the method.

[0153] Figure 7 It is further shown that in the large aberration In this case, the effect of image reconstruction, despite the large aberration, can still obtain satisfactory reconstruction results through proper adjustment.

[0154] Figure 8 Figure (a) is the spectrum graph displayed by the detector data collection, Figure (b) is the spectrum result graph after adaptive optics correction, Figure (c) is the spectrum result graph after restoration by the method of the present invention, and Figure (d) is a spectrum comparison graph of the adaptive optics correction and the restoration results of the method of the present invention. Figure 8 As can be seen from the figure, the correction effect of the present invention is comparable to that of adaptive optics, demonstrating the accuracy and effectiveness of the present invention. Compared to traditional adaptive optics, the present invention can still achieve high-performance imaging capabilities with a large field of view under conditions of large aberrations and noise.

[0155] In summary, imaging grating spectrometers not only provide powerful spectral analysis capabilities, but also significantly enhance system performance and application range through advanced aberration correction techniques and image processing algorithms. The goal of this technical solution is to provide an efficient and accurate method to overcome image degradation caused by optical system limitations and physical conditions, thereby promoting the development of high-spectral-resolution grating spectrometer technology.

[0156] The above is a detailed introduction to the slit-type spectral imaging method based on filtered phase reconstruction and the grating spectrometer provided by the present invention. In this embodiment, specific examples are used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for general technical personnel in this field, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

[0157] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined in this embodiment may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown in this embodiment, but is intended to conform to the widest scope consistent with the principles and novel features disclosed in this embodiment.

Claims

1. A slit grating spectral imaging method based on filtered phase reconstruction, characterized in that: The steps include: S1: Using a wavefront sensor to measure the wavefront aberration before slit filtering in a slit grating imaging spectrometer; S2: Based on the slit filtering principle, the wavefront aberration before slit filtering is used to reconstruct the phase function after slit filtering; S3: Capturing a slit image, and reconstructing the slit image using the reconstructed phase function after slit filtering to obtain a final reconstructed image.

2. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 1, characterized in that: The S1 comprises the following steps: S11: Split the incident beam into multiple small wavefronts using a microlens array; S12: measuring the focus offset before extracting each wavelet; S13: Calculating the wavefront slope according to the center coordinates of each microlens and the focus offset of its corresponding wavelet; S14: The wavefront phase function is converted into a Zernike polynomial expansion form using the wavefront slope, and the polynomial coefficients are fitted using the least squares method. The wavefront aberration before slit filtering is quantified based on the polynomial coefficients.

3. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 2, characterized in that: The Zernike polynomial expansion form is: Where φ(x,y) is the wavefront phase function, a n is the coefficient of the polynomial to be determined, Z n (x,y) is the nth order Zernike polynomial, (x,y) is the focal coordinate of the wavelet front; Use the least squares method to fit the polynomial coefficient a to be determined n , such that: Where s x (x i ,y i ),s y (x i ,y i ) is the coordinate of the i-th wavelet front intersection point (x i ,y i ) corresponds to the x-axis wavefront slope and y-axis wavefront slope.

4. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 1, characterized in that: The S2 comprises the following steps: S21: Calculate the near-field function E of the optical path before slit filtering based on the wavefront aberration before slit filtering nf (x,y); S22: through the near-field function E nf The far-field function E of the optical path before slit filtering is obtained by Fourier transform of (x,y) far (x,y); S23: Through the far-field function E far The Fourier transform of (x,y) reconstructs the phase function after slit filtering.

5. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 4, characterized in that: The phase function after slit filtering reconstructed by S23 Expressed as: Where, the functions real{} and angle{} return the real part and phase angle of the complex array respectively, F{} represents the Fourier transform, pupil(x,y) represents the pupil aperture function, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, i is the imaginary unit, w s is the width of the slit, and Ls is the length of the slit.

6. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 1, characterized in that: The S3 comprises the following steps: S31: Capture the slit image and use it as the blurred image O(x,y); S32: Reconstruct the point spread function (PSF) based on the phase function after slit filtering; S33: Iteratively update the estimated value I of the blurred image O(x,y) using the reconstructed point spread function PSF n+1 (x, y), which is the final reconstructed image.

7. The slit grating spectral imaging method based on filtered phase reconstruction according to claim 6, characterized in that: The S33 uses the LRRA algorithm to reconstruct the image, which is expressed as: Where A0 is the initial value or reference amplitude of the wavefront amplitude, is the complex amplitude distribution of the wavefront, including the wavefront aberration before filtering, and ⊙ is the nonlinear reconstruction operator, defined as: in, is the Fourier transform of A, F -1 is the inverse Fourier transform, and α and β are adjusted between -1 and 1 with the goal of minimizing noise.

8. A slit grating spectrometer using a slit grating spectral imaging method based on filtered phase reconstruction according to any one of claims 1 to 7, characterized in that: include: A laser light source module and a beam splitter are sequentially arranged along the propagation direction of the optical path. The beam splitter splits the incident light into two paths. The first optical path generates a preset phase distribution through an aberration generator and is reflected and fused to the second optical path. The wavefront sensor measures the wavefront aberration in the second optical path after being modulated by the aberration generator and before being slit filtered. After passing through the slit, the second light path is collimated by an off-axis parabolic mirror and projected onto a reflective grating; the reflective grating disperses the light in different directions according to different wavelengths to form a spectral image; the detector captures the spectral image and sends it to the image processing and algorithm module to reconstruct the slit image to obtain the final reconstructed image.

9. The slit grating spectrometer according to claim 8, characterized in that: The laser light source module adopts a monochromatic laser.

10. The slit grating spectrometer according to claim 8, characterized in that: The laser light source module and the beam splitter are further provided in sequence along the light path propagation direction: Pinhole, used to filter out stray light in the outgoing beam of the laser light source module; The collimator is used to collimate the light beam after passing through the pinhole, forming parallel light and then incident on the beam splitter.

Images