A laminated diffraction position correction imaging method, system and electronic device

Abstract

The present application belongs to the technical field of coherent diffraction imaging, and discloses a kind of laminated diffraction position correction imaging method, system and electronic equipment, wherein the method comprises: for each scan position coordinate of motion platform, the minimum difference between the calculated diffraction light field intensity information and the collected diffraction light field intensity information is the optimal value by particle swarm algorithm, and the updated integer pixel position coordinate of motion platform is obtained by iterative updating;The cross-correlation peak position before and after the diffraction exit wave update of each updated integer pixel position coordinate of motion platform is calculated, and the updated sub-pixel position coordinate of motion platform is obtained by continuously searching and updating the irradiation probe position;The complex amplitude function of irradiation probe and sample to be measured is calculated, and the amplitude and phase of sample to be measured and irradiation probe are reconstructed from this.The convergence speed of position correction algorithm is improved by effectively combining particle swarm position correction method and cross-correlation position correction method, and the final accuracy is also guaranteed.

Description

Technical Field

[0001] This invention belongs to the technical field of coherent diffraction imaging, and more specifically, relates to a method, system and electronic device for stacked diffraction position correction imaging. Background Technology

[0002] As science and technology continue to advance, the demands for image quality are also increasing. Coherent diffractive imaging (CDI), as a lensless imaging technique, can effectively overcome the errors caused by traditional microscope lens groups. However, during the experiment, it only records diffraction intensity information, losing the phase information of the sample itself. Nevertheless, as long as the illumination probe meets certain coherence conditions, the phase information of the sample can be recovered through refined sampling. The phase recovery problem has become the core of computational imaging.

[0003] Ptychography, an important branch of CDI technology, relies on the core idea that adjacent regions scanned by the illumination probe must have a certain overlap rate. The consistency of the complex amplitude function of the sample under test within the overlapping region forms a constraint, enabling the phase retrieval algorithm to converge quickly. Theoretically, iterative projection algorithms can solve the phase retrieval problem. A major advantage of ptychography in reconstructing the amplitude and phase information of the sample under test is the ability to obtain a large field of view. The relative movement between the illumination probe and the sample under test is the condition for generating a large field of view, and the accuracy of the illumination probe's movement is directly related to the reconstruction of the amplitude and phase information of the sample under test. Therefore, positional errors have a profound impact on image reconstruction.

[0004] In recent years, some achievements have been made both domestically and internationally in the field of position correction algorithms for stacked diffraction. For example, in 2012, AM Maiden et al. applied the idea of ​​simulated annealing algorithm to Ptychography position correction, exploring the optimal position of the probe by randomly searching the position of the illumination probe. However, existing position correction techniques in stacked diffraction imaging often suffer from decreased computational efficiency as computational accuracy increases, presenting a technical problem of not being able to simultaneously improve accuracy and speed. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a stacked diffraction position correction imaging method, system and electronic device, which solves the technical problem that the position correction technology in the existing stacked diffraction imaging cannot simultaneously improve accuracy and speed.

[0006] To achieve the above objectives, according to a first aspect of the present invention, a method for correcting the position of stacked diffraction is provided, wherein during the process of the sample under test moving in a two-dimensional plane according to a preset curve driven by a motion platform, an illumination probe is coupled to the sample under test, and the intensity information of the diffraction light field of the sample under test is acquired by a camera. The method includes:

[0007] For each scanning position coordinate of the motion platform, a series of position coordinates are randomly generated with that coordinate as the center. The series of position coordinates are divided into multiple populations. The optimal value is the one with the smallest difference between the diffraction light field intensity information calculated for each position coordinate in the population and the diffraction light field intensity information acquired by the camera. The optimal solution of each population is iteratively updated to obtain the updated integer pixel position coordinates of the motion platform.

[0008] Based on the updated integer pixel position coordinates of the motion platform, the cross-correlation peak positions of the diffraction exit wave before and after the update of each integer pixel position coordinate of the motion platform are calculated. The position of the illumination probe is continuously searched and updated for iterative optimization to obtain the updated sub-pixel position coordinates of the motion platform.

[0009] Based on the updated subpixel position coordinates of the motion platform, and the complex amplitude functions of the irradiation probe and the sample under test calculated corresponding to the subpixel position coordinates, the amplitude and phase of the sample under test and the irradiation probe are reconstructed.

[0010] According to the stacked diffraction position correction imaging method provided by the present invention, S1 obtaining the updated integer pixel position coordinates of the motion platform specifically includes:

[0011] For each scan position coordinate of the motion platform, a series of position coordinates are randomly generated with the initial position coordinates as the center, and the series of position coordinates are divided into multiple populations;

[0012] For any population, calculate the complex amplitude function of the exit wave illuminating the probe at each position coordinate in the population; propagate the complex amplitude function of the exit wave to the reciprocal space to calculate the diffraction field intensity information; take the position coordinate with the smallest difference between the calculated diffraction field intensity information and the diffraction field intensity information acquired by the camera as the individual's historical best position for that population; at the same time, take the individual's historical best position with the smallest difference among multiple populations as the population's historical best position.

[0013] The weighted sum of the distance of position update in the previous iteration, the difference between the individual historical best position and the position of the motion platform, and the difference between the population historical best position and the position of the motion platform is used as the distance of position update in the next iteration. With the goal of minimizing the difference between the calculated diffraction light field intensity information and the diffraction light field intensity information collected by the camera, the population is iteratively updated. The integer pixel position coordinates of the updated motion platform are obtained by adding the distance of position update corresponding to the final iteration number to the position coordinates of the motion platform corresponding to the previous iteration.

[0014] According to the stacked diffraction position correction imaging method provided by the present invention, the distance v of position update in the next iteration and the integer pixel position coordinates r of the updated motion platform ′ are calculated by the following method:

[0015] v = v0 * w + c1 * (x m - r) + c2 * (y m - r);

[0016] r ′ = r + v;

[0017] where, w represents the position inertia weight in the previous iteration, and its value is between [0, 1]; c1 represents the self-update weight, and its value is between [0, 1]; c2 represents the population update weight, and its value is between [0, 1]; x m represents the average value of the individual historical best positions; y m represents the population historical best position; v0 is the distance of position update in the previous iteration; r represents the position coordinates of the motion platform; r ′ represents the position coordinates of the updated motion platform.

[0018] According to the stacked diffraction position correction imaging method provided by the present invention, S1 further includes:

[0019] Iteratively update the optimal solutions of each population until the root mean square error between the calculated diffraction light field intensity information and the diffraction light field intensity information collected by the camera reaches the first preset threshold.

[0020] According to the stacked diffraction position correction imaging method provided by the present invention, specifically, when the root mean square error MSE < A between the calculated diffraction light field intensity information and the diffraction light field intensity information collected by the camera, the iteration is completed;

[0021]

[0022] where, I m (q) represents the diffraction light field intensity information collected by the camera at the mth scanning position; Φ m(q) represents the guessed complex amplitude function of the coherent diffraction field at the m-th scanning position; A represents the first preset threshold, which is an empirical value.

[0023] According to the stacked diffraction position correction imaging method provided by the present invention, in S2, the cross-correlation peak positions of the diffraction exit wave before and after the update of each integer pixel position coordinate of the motion platform are calculated, specifically in the following manner:

[0024] CC m =∑O m ′ (r)·P m ′ (r,λ)·P m (r,λ)·O m (r);

[0025] Among them, CC m The correlation matrix in the probe; O m ′ (r) is the complex amplitude function of the sample after updating at the m-th scan position; P m ′ (r,λ) is the complex amplitude function of the irradiated probe after the m-th scan position is updated; P m (r,λ) is the complex amplitude function of the illumination probe corresponding to the previous iteration at the m-th scanning position; O m (r) is the complex amplitude function of the sample to be tested corresponding to the previous iteration at the m-th scanning position.

[0026] According to the stacked diffraction position correction imaging method provided by the present invention, the iterative optimization of searching and updating the illumination probe position in S2 is specifically calculated in the following manner:

[0027] D j =find(|CC m |==max(|CC m |);

[0028] D j ′ =θD j ;

[0029] r ′ =r+D j ′ ;

[0030] Among them, D j The value of the probe position correction for the j-th irradiation; r represents the position coordinates of the motion platform; r ′ The position coordinates of the motion platform after the update are represented; find represents a function that finds a certain condition in the numerical matrix; θ represents the adjustment coefficient of the irradiation probe position correction amount.

[0031] According to the stacked diffraction position correction imaging method provided by the present invention, the complex amplitude function of the sample under test is:

[0032]

[0033] The complex amplitude function of the irradiation probe is:

[0034]

[0035] Where α represents the first iteration search step size, with a value between [0,1]; β represents the second iteration search step size, with a value between [0,1]; O m (r), O m ′ (r) represents the complex amplitude function of the sample before and after the position coordinate update at the m-th scan position, respectively; P m (r,λ), P m ′ (r,λ) represent the complex amplitude functions of the irradiation probe before and after the position coordinate update at the m-th scanning position, respectively; These represent the diffraction light field intensity information before and after the amplitude replacement of the illumination probe at the m-th scanning position, respectively; r represents the position coordinates of the motion platform; λ represents the wavelength of the illumination probe; * represents the conjugate operation; || max This represents the maximum value of the amplitude of each element in the matrix corresponding to the complex amplitude function.

[0036] According to a second aspect of the present invention, a stacked diffraction position correction imaging system is provided, comprising a laser, a beam expander, an aperture, a focusing lens, a sample to be tested, and a camera arranged sequentially along the optical path direction. The laser passes sequentially through the beam expander, the aperture, and the focusing lens to generate an illumination probe that illuminates the sample to be tested on a motion platform. The system also includes a processor for executing the stacked diffraction position correction imaging method described in any of the preceding claims.

[0037] According to a third aspect of the invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the stacked diffraction position correction imaging method as described in any of the preceding claims.

[0038] In summary, compared with the prior art, the stacked diffraction position correction imaging method, system, and electronic device provided by the present invention offer the following advantages:

[0039] 1. By effectively combining the particle swarm optimization (PSO) position correction method with the cross-correlation position correction method, the PSO algorithm is used to rapidly iterate and update the search position for global fast position correction at the integer pixel level, while the cross-correlation algorithm is used for fine local sub-pixel level correction. Compared with other stacked diffraction computational imaging methods, this method improves the convergence speed of the position correction algorithm while ensuring the final accuracy.

[0040] 2. The stacked diffraction position correction imaging method based on particle swarm optimization can quickly and effectively search for the direction of position correction, and can also avoid the algorithm getting stuck in local optima when the error is large. Thus, through the whole pixel level correction process, it can quickly reduce the global position error of the illumination probe position coordinates and improve the correction speed.

[0041] 3. Based on the cross-correlation stacked diffraction position correction algorithm, fine correction of position errors at the integer pixel level can greatly improve the quality of the reconstructed amplitude and phase image, overcome the degradation of image local detail reconstruction quality caused by positional disturbance, avoid the problem of sample sharpness degradation in stacked diffraction computational imaging, improve the resolution of the reconstructed image, and ensure imaging accuracy. Attached Figure Description

[0042] Figure 1 This is a schematic flowchart of the stacked diffraction position correction imaging method provided by the present invention;

[0043] Figure 2 This is a flowchart of a method for correcting the position of stacked diffraction provided by the present invention;

[0044] Figure 3 This is a schematic diagram of the optical path of the stacked diffraction position correction imaging system provided by the present invention;

[0045] Figure 4 (a) is the amplitude pattern of the sample under test used in the simulation process provided by the present invention;

[0046] Figure 4 (b) is the phase pattern of the sample under test used in the simulation process provided by the present invention;

[0047] Figure 5 This invention provides the two-dimensional motion plane trajectory of the sample to be tested;

[0048] Figure 6 (a) and (b) are the sample amplitude reconstruction diagram and phase reconstruction diagram without position correction provided by the present invention;

[0049] Figure 6 Images (c) and (d) are the pixel-level amplitude reconstruction images and phase reconstruction images provided by the present invention after particle swarm optimization algorithm-based stacked diffraction position correction.

[0050] Figure 6 Images (e) and (f) are the amplitude reconstruction images and phase reconstruction images provided by the present invention after sub-pixel level correction of cross-correlation stacked diffraction position correction;

[0051] Figure 7 (a) is a distribution diagram of the irradiation probes provided by the present invention without position correction;

[0052] Figure 7 (b) is the irradiation probe distribution diagram provided by the present invention after pixel-level correction by the particle swarm algorithm-based stacked diffraction position correction method;

[0053] Figure 7 (c) is a diagram of the irradiation probe distribution after sub-pixel level correction by the cross-correlation stacked diffraction position correction method provided by the present invention;

[0054] Figure 8 This is a schematic diagram of the electronic device provided by the present invention;

[0055] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein:

[0056] 1 is the laser, 2 is the mirror, 3 is the beam expander, 4 is the aperture, 5 is the converging lens, 6 is the sample to be tested, and 7 is the camera. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0058] Please see Figure 1 This invention provides a method for position correction imaging in stacked diffraction. The method is applied to a stacked diffraction system, which includes a laser 1, a beam expander 3, an aperture 4, a focusing lens, a sample under test, and a camera 7 arranged sequentially along the optical path. The laser 1, through the beam expander 3, the aperture 4, and the focusing lens, generates an illumination probe that illuminates the sample under test on a moving platform. During the two-dimensional planar motion of the sample under test along a preset curve driven by the moving platform, the illumination probe and the sample under test are coupled, and the camera 7 acquires the intensity information of the diffraction field of the sample under test. The method includes:

[0059] For each scanning position coordinate of the motion platform, a series of position coordinates are randomly generated with it as the center. The series of position coordinates are divided into multiple populations. The optimal value is the one with the smallest difference between the diffraction light field intensity information calculated by each position coordinate in the population and the diffraction light field intensity information collected by the camera 7. The optimal solution of each population is iteratively updated to obtain the updated integer pixel position coordinates of the motion platform.

[0060] Based on the updated integer pixel position coordinates of the motion platform, the cross-correlation peak positions of the diffraction exit wave before and after the update of each integer pixel position coordinate of the motion platform are calculated. The position of the illumination probe is continuously searched and updated for iterative optimization to obtain the updated sub-pixel position coordinates of the motion platform.

[0061] Based on the updated subpixel position coordinates of the motion platform, and the complex amplitude functions of the irradiation probe and the sample under test calculated corresponding to the subpixel position coordinates, the amplitude and phase of the sample under test and the irradiation probe are reconstructed.

[0062] The stacked diffraction position correction imaging method provided by this invention considers the motion error of the sample under test in the stacked diffraction system, which leads to inaccurate position of the sample and reduces the amplitude and phase quality of the reconstructed sample. This invention proposes that when the sample under test moves along a preset trajectory with the motion platform, a stacked diffraction position correction algorithm based on particle swarm optimization (PSO) is first used to iteratively update the search position for rapid integer-pixel-level position correction. The PSO-based method can quickly and effectively search for the direction of position correction and avoid the algorithm getting trapped in local optima when the error is large. Thus, through the integer-pixel-level correction process, the global position error of the illumination probe coordinates can be quickly reduced, improving the correction speed. Then, a cross-correlation-based stacked diffraction position correction algorithm is used to finely correct the integer-pixel-level position error, which can greatly improve the quality of the reconstructed amplitude and phase image. This overcomes the degradation of local image detail reconstruction quality caused by positional perturbations, avoids the problem of sample sharpness degradation in stacked diffraction computational imaging, improves the resolution of the reconstructed image, and ensures imaging accuracy.

[0063] In some specific embodiments, S1 obtaining the updated integer pixel position coordinates of the motion platform specifically includes:

[0064] For each scan position coordinate of the motion platform, a series of position coordinates are randomly generated with the initial position coordinates as the center, and the series of position coordinates are divided into multiple populations;

[0065] For any population, calculate the exit wave complex amplitude function of the probe at each position coordinate in the population; propagate the exit wave complex amplitude function to the reciprocal space to calculate the diffraction field intensity information; take the position coordinate with the smallest difference between the calculated diffraction field intensity information and the diffraction field intensity information acquired by camera 7 as the individual's historical best position for that population; simultaneously, take the individual's historical best position with the smallest difference among multiple populations as the group's historical best position; that is, for the i-th population, calculate the exit wave complex amplitude function of the probe at each particle's position coordinate in the population; propagate the exit wave complex amplitude function to the reciprocal space to calculate the diffraction field intensity information; take the position coordinate with the smallest difference between the calculated diffraction field intensity value and the diffraction field intensity value acquired by camera 7 as the individual's historical best position for the i-th population; simultaneously, take the individual's historical best position with the smallest difference among multiple populations as the group's historical best position.

[0066] The weighted sum of the distance updated in the previous iteration, the difference between the individual's historical best position and the position of the motion platform, and the difference between the group's historical best position and the position of the motion platform is used as the distance updated in the next iteration. The population is iteratively updated with the goal of minimizing the difference between the calculated diffraction light field intensity information and the diffraction light field intensity information acquired by camera 7. The distance updated in the final iteration is added to the position coordinates of the motion platform corresponding to the previous iteration to obtain the updated integer pixel position coordinates of the motion platform.

[0067] Specifically, the distance v updated in the next iteration and the integer pixel position coordinates r updated by the motion platform. ′ Calculated as follows:

[0068] v = v0 * w + c1 * (x m -r)+c2*(y m -r);

[0069] r ′ =r+v;

[0070] Where w represents the position inertia weight from the previous iteration, with a value between [0,1]; c1 represents the self-update weight, with a value between [0,1]; c2 represents the group update weight, with a value between [0,1]; x m y represents the average of an individual's historical best positions; m The group's historical best position is represented by v0, which is the distance from the position update in the last iteration; r represents the position coordinates of the motion platform; r ′ This indicates the updated position coordinates of the motion platform.

[0071] In some specific embodiments, S1 further includes:

[0072] Iteratively update the optimal solutions of each population until the root mean square error between the calculated diffracted light field intensity information and the diffracted light field intensity information collected by Camera 7 reaches the first preset threshold.

[0073] That is, when performing integer pixel-level position correction on the illumination probe based on the particle swarm optimization algorithm, iteratively update the optimal solutions of each population until the MSE parameter reaches the threshold, and obtain the integer pixel position coordinates of the updated illumination probe.

[0074] In some specific embodiments, specifically, according to the illumination probe after integer pixel correction, when the root mean square error MSE between the calculated diffracted light field intensity information and the diffracted light field intensity information collected by Camera 7 is less than A, the iteration is completed;

[0075]

[0076] Among them, I m (q) represents the diffracted light field intensity information collected by Camera 7 at the m-th scanning position; Φ m (q) represents the guessed complex amplitude function of the coherent diffracted light field at the m-th scanning position; A represents the first preset threshold, which is an empirical value. That is, when the root mean square error is less than a certain threshold A, the difference between the diffracted light field calculated from the reconstructed probe sample and the diffracted light field intensity value collected by Camera 7 is small. A is an empirical value, and in this embodiment, A is taken as 0.1; when MSE < 0.1, the integer pixel level of the iterative diffraction position correction algorithm based on the particle swarm reaches a convergence state. The present invention proposes that when the reconstructed amplitude-phase image MSE < 0.1 in the iterative diffraction position correction based on the particle swarm, the global position error of the illumination probe is at a low level.

[0077] In some specific embodiments, in S2, calculate the cross-correlation peak positions before and after the update of the diffracted exit wave at each integer pixel position coordinate of the updated moving platform, and specifically calculate through the following method:

[0078] CC m =∑O m ′ (r)·P m ′ (r,λ)·P m (r,λ)·O m (r);

[0079] Among them, CC m is the correlation numerical matrix in the probe; O m ′ (r) is the complex amplitude function of the待测 sample after update at the m-th scanning position; P m ′ (r,λ) is the complex amplitude function of the illumination probe after update at the m-th scanning position; Pm (r,λ) is the complex amplitude function of the illumination probe corresponding to the previous iteration at the m-th scanning position; O m (r) is the complex amplitude function of the sample to be tested corresponding to the previous iteration at the m-th scan position. That is, O m (r), O m ′ (r) represents the complex amplitude function of the sample to be tested corresponding to two adjacent iterations at the m-th scanning position; P m (r,λ), P m ′ (r,λ) is the complex amplitude function of the irradiation probe corresponding to two adjacent iterations at the m-th scanning position.

[0080] In some specific embodiments, the iterative optimization of searching and updating the irradiation probe position in S2 is specifically calculated in the following way:

[0081] D j =find(|CC m |==max(|CC m |);

[0082] D j ′ =θD j ;

[0083] r ′ =r+D j ′ ;

[0084] Among them, D j The value of the probe position correction for the j-th irradiation; r represents the position coordinates of the motion platform; r ′ The updated position coordinates of the motion platform are represented by ; find represents a function that finds a certain condition in the numerical matrix; θ represents the adjustment coefficient of the illumination probe position correction quantity. The updated illumination probe position coordinates after integer-pixel correction are input into a cross-correlation-based stacked diffraction position correction algorithm for sub-pixel correction until a preset maximum number of iterations is reached or the MSE parameter reaches a threshold, thus obtaining the updated sub-pixel position correction coordinates of the illumination probe.

[0085] Subpixel correction specifically involves inputting the coordinates of the illumination probe after integer pixel correction, calculating the cross-correlation peak positions of the diffraction exit waves before and after updating for each illumination probe position, continuously searching and updating the illumination probe positions until the preset maximum number of iterations is reached or the MSE parameter reaches a threshold, thus obtaining the subpixel position correction after the illumination probe update.

[0086] In some specific embodiments, the complex amplitude function of the sample to be tested is:

[0087]

[0088] The complex amplitude function of the irradiation probe is:

[0089]

[0090] Where α represents the first iteration search step size, with a value between [0,1]; β represents the second iteration search step size, with a value between [0,1]; O m (r), O m ′ (r) represents the complex amplitude function of the sample before and after the position coordinate update at the m-th scan position, respectively; P m (r,λ), P m ′ (r,λ) represent the complex amplitude functions of the irradiation probe before and after the position coordinate update at the m-th scanning position, respectively; These represent the diffraction light field intensity information before and after the amplitude replacement of the illumination probe at the m-th scanning position, respectively; r represents the position coordinates of the motion platform; λ represents the wavelength of the illumination probe; * represents the conjugate operation; || max This represents the maximum value of the amplitude of each element in the matrix corresponding to the complex amplitude function.

[0091] The present invention also provides a stacked diffraction position correction imaging system, which includes a laser 1, a beam expander 3, an aperture 4, a focusing lens, a sample to be tested, and a camera 7 arranged sequentially along the optical path. The laser 1 passes through the beam expander 3, the aperture 4, and the focusing lens in sequence to generate an illumination probe that illuminates the sample to be tested on the motion platform. The system also includes a processor, which is used to execute the stacked diffraction position correction imaging method described in any of the above claims.

[0092] Specific Example 1

[0093] A method for correcting the position of a stacked diffraction system, the method being applied to a stacked diffraction system, with reference to... Figure 3 The stacked diffraction system includes a laser 1, a mirror 2, a beam expander 3, an aperture 4, a converging lens 5, a sample 6, and a camera 7 (CCD or CMOS) arranged sequentially along the optical axis. The working wavelength of the helium-neon laser 1 is 632.8 nm, and the beam diameter (1 / e2) of the output of the helium-neon laser 1 is 0.54 mm. After being reflected by the mirror 2, the beam expander 3 expands the beam by 10 times. The aperture 4 adjusts the spot diameter to 2 mm. The collimated parallel beam passes through a focusing lens with an effective focal length of 10 cm and then illuminates the sample 6 located about 1.5 mm away from the back focal plane of the focusing mirror. This generates an illumination probe, i.e., an illumination probe, which illuminates the sample 6 on the moving platform. Finally, the camera 7 collects the diffracted light field.

[0094] refer to Figure 2 The method includes:

[0095] A precision motion platform drives the sample 6 to perform two-dimensional planar motion according to a specific preset curve (including but not limited to grid curves), ensuring that the overlap area of ​​adjacent illumination spots is not less than 60% as a spatial overlap constraint. The illumination probe illuminates the sample 6, and a camera 7 collects and records a series of diffraction field intensity information. m (q), where m is the scanning position during the motion, and q is the frequency domain coordinate;

[0096] This method employs a particle swarm optimization-based stacked diffraction position correction approach to rapidly correct global errors at the integer pixel level. Simultaneously, the correction status is assessed using the image's MSE (Mean Sequence Equation) index. Finally, a cross-correlation-based stacked diffraction position correction approach is used to refine local errors to the sub-pixel level. The complex amplitude functions of the illumination probe and the sample under test are then calculated, thereby reconstructing the amplitude and phase of the sample. This approach combines two stacked diffraction position correction algorithms, achieving high accuracy while maintaining a fast convergence speed.

[0097] Specifically, firstly, regarding the complex amplitude function P of the illumination probe... m (r,λ), the complex amplitude function O of the sample to be tested m (r), providing an initial guess. The position coordinates before the offset are set as the position coordinates r of the motion platform; motion scanning is performed according to the preset scanning path, and the illumination probe P is calculated at different scanning positions. m (r,λ) and the sample O to be tested m (r) Complex amplitude function of the exit wave formed by the interaction Export wave complex amplitude function By propagating to Fourier space, i.e., reciprocal space, the intensity information of the diffracted light field at different scanning positions of the detection plane of camera 7 is obtained; amplitude substitution is performed in reciprocal space, and the complex amplitude update function of the exit wave at different scanning positions can be calculated through the inverse propagation model. Based on the complex amplitude functions of the exit wave before and after the update, the complex amplitude functions of the illumination probe and the sample under test are updated simultaneously using the update function. Then, the position coordinates of the illumination probe are updated using the particle swarm optimization algorithm and the cross-correlation algorithm. The updated position coordinates of the sample under test and the illumination probe are output until the error is small enough to reach a preset threshold.

[0098] Specific Example 2

[0099] A stacked diffraction imaging system, such as Figure 3As shown, this system is a transmission imaging system, including a laser 1, a reflector 2, a beam expander 3, an aperture 4, a converging lens 5, a sample under test 6, and a camera 7 (CCD or CMOS) arranged sequentially along the optical axis; it also includes a processor, which is used to execute the stacked diffraction position correction imaging method described above. The sample under test has motion errors due to the motion platform, and the processor is used to correct the coordinates of the illumination probe and the sample under test on the motion platform.

[0100] Figure 4 In the middle (a) and (b), respectively, are the amplitude and phase patterns of the sample under test used in the simulation process; the illumination probe wavelength is 632.8 nm, the sample size is 256×256 pixels, and the beam diameter is 64×64 pixels of Gaussian beam.

[0101] Figure 5 This is the two-dimensional motion plane trajectory of the sample under test provided in this embodiment of the invention. The precision motion stage drives the sample under test to perform two-dimensional planar motion, with a motion step size of 5 pixels in the XY direction, and the sample under test is scanned in a 10×10 area. Therefore, the CMOS camera 7 collects a total of 100 coherent diffraction fields with a position error of ±10 pixels.

[0102] Figure 6 In the middle (a) and (b), the amplitude and phase patterns of the sample without position correction are shown respectively. Figure 6 Images (c) and (d) are, respectively, the amplitude and phase patterns of integer-pixel correction samples based on particle swarm optimization (PSO) algorithm for stacked diffraction position correction. Figure 6 In the middle (e) and (f), respectively, are the amplitude and phase patterns of the subpixel-corrected samples after cross-correlation stacked diffraction position correction method.

[0103] Figure 7 In the middle, (a), (b), and (c) are respectively the illumination probe distribution map without position correction, the illumination probe distribution map after integer pixel-level correction by the particle swarm algorithm stacked diffraction position correction method, and the illumination probe distribution map after sub-pixel-level correction by the cross-correlation stacked diffraction position correction method.

[0104] contrast Figure 6 As can be seen from the figures, the reconstructed sample amplitude image without position correction has a low resolution because the image reconstruction quality is reduced due to positional errors. This method overcomes the image reconstruction quality degradation caused by positional perturbations and effectively improves the resolution of the reconstructed image.

[0105] This invention proposes a stacked diffraction position correction method. By effectively combining particle swarm optimization (PSO) and cross-correlation (CCO) position correction methods, it employs a PSO algorithm for rapid iterative updates of the search position for global position correction at the integer pixel level, and a CCO algorithm for refined local sub-pixel level correction. Compared to other stacked diffraction computational imaging methods, this method improves the convergence speed of the position correction algorithm while ensuring final accuracy.

[0106] Furthermore, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the stacked diffraction position correction imaging method as described in any of the above embodiments.

[0107] Furthermore, a non-transitory computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the stacked diffraction position correction imaging method as described in any of the above embodiments.

[0108] Figure 8 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 8 As shown, the electronic device may include a processor, a communications interface, a memory, and a communication bus, wherein the processor, communications interface, and memory communicate with each other via the communication bus. The processor can call logical instructions in the memory to execute the stacked diffraction position correction imaging method.

[0109] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0110] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, and when the program instructions are executed by a computer, the computer is able to execute the stacked diffraction position correction imaging method provided by the above methods.

[0111] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the stacked diffraction position correction imaging method provided by the methods described above.

[0112] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0113] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0114] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for stacked diffraction position correction imaging, wherein during the process of the sample under test moving in a two-dimensional plane according to a preset curve driven by a motion platform, an illumination probe is coupled to the sample under test, and the intensity information of the diffraction light field of the sample under test is acquired by a camera, characterized in that, The method includes: S1. For each scanning position coordinate of the motion platform, a series of position coordinates are randomly generated with it as the center. The series of position coordinates are divided into multiple populations. The optimal value is the one with the smallest difference between the diffraction light field intensity information calculated for each position coordinate in the population and the diffraction light field intensity information collected by the camera. The optimal solution of each population is iteratively updated to obtain the updated integer pixel position coordinates of the motion platform. S2, based on the updated integer pixel position coordinates of the motion platform, calculate the cross-correlation peak positions of the diffraction exit wave before and after the update of each integer pixel position coordinate of the motion platform, continuously search and update the illumination probe position for iterative optimization, and obtain the updated sub-pixel position coordinates of the motion platform. S3, based on the updated sub-pixel position coordinates of the motion platform, and the complex amplitude functions of the irradiation probe and the sample under test calculated corresponding to the sub-pixel position coordinates, the amplitude and phase of the sample under test and the irradiation probe are reconstructed. S1 also includes: The optimal solution for each population is iteratively updated until the root mean square error between the calculated diffraction light field intensity information and the diffraction light field intensity information acquired by the camera reaches the first preset threshold. The root mean square error between the calculated diffraction light field intensity information and the diffraction light field intensity information acquired by the camera. When the iteration is complete; ； in, This represents the intensity information of the diffraction light field acquired by the camera at the m-th scanning position; This represents the guessed complex amplitude function of the coherent diffraction field at the m-th scan position; A represents the first preset threshold, which is an empirical value. In S2, the cross-correlation peak positions of the diffraction exit wave before and after the update of each integer pixel position coordinate of the motion platform are calculated, specifically in the following manner: ； in, This is the correlation matrix of the probe; The complex amplitude function of the sample to be tested after updating at the m-th scan position; The complex amplitude function of the illuminated probe is updated at the m-th scan position; Let be the complex amplitude function of the irradiation probe corresponding to the previous iteration at the m-th scanning position; Let be the complex amplitude function of the sample to be tested corresponding to the previous iteration at the m-th scanning position; The iterative optimization of searching and updating the illumination probe position in S2 is specifically calculated in the following way: ； ； ； in, This is the probe position correction amount for the j-th irradiation; Indicates the position coordinates of the motion platform; This indicates the updated position coordinates of the motion platform; A function representing a condition found in a numerical matrix; This represents the adjustment coefficient for the correction of the irradiation probe position.

2. The stacked diffraction position correction imaging method as described in claim 1, characterized in that, S1 obtains the updated integer pixel position coordinates of the motion platform, specifically including: For each scan position coordinate of the motion platform, a series of position coordinates are randomly generated with the initial position coordinates as the center, and the series of position coordinates are divided into multiple populations; For any population, calculate the complex amplitude function of the exit wave illuminating the probe at each position coordinate in the population; propagate the complex amplitude function of the exit wave to the reciprocal space to calculate the diffraction field intensity information; take the position coordinate with the smallest difference between the calculated diffraction field intensity information and the diffraction field intensity information acquired by the camera as the individual's historical best position for that population; at the same time, take the individual's historical best position with the smallest difference among multiple populations as the population's historical best position. The weighted sum of the distance updated in the previous iteration, the difference between the individual's historical best position and the position of the motion platform, and the difference between the group's historical best position and the position of the motion platform is used as the distance updated in the next iteration. The population is iteratively updated with the goal of minimizing the difference between the calculated diffraction light field intensity information and the diffraction light field intensity information acquired by the camera. The distance updated in the final iteration is added to the position coordinates of the motion platform corresponding to the previous iteration to obtain the updated integer pixel position coordinates of the motion platform.

3. The stacked diffraction position correction imaging method as described in claim 2, characterized in that, The distance of the position update in the next iteration and the updated integer pixel position coordinates of the motion platform Calculated as follows: ； ； in, This represents the position inertia weight from the previous iteration, and its value is between [0,1]. This represents the self-updating weight, with a value between [0,1]. This represents the group update weight, with a value between [0,1]. This represents the average of an individual's historical best positions. Indicates the group's best historical position; This is the distance from the position update in the previous iteration.

4. The stacked diffraction position correction imaging method as described in claim 1, characterized in that, The complex amplitude function of the sample after updating at the m-th scan position is: ； The complex amplitude function of the illumination probe after updating at the m-th scan position is: ； in, This represents the search step size for the first iteration, and its value is between [0,1]. This represents the search step size for the second iteration, and its value is between [0,1]. They represent the first Information on the intensity of the diffraction field before and after the amplitude of the irradiated probe at each scanning position; Indicates the wavelength of the irradiation probe; * indicates conjugate operation; This represents the maximum value of the amplitude of each element in the matrix corresponding to the complex amplitude function.

5. A stacked diffraction position correction imaging system, characterized in that, The device includes a laser, a beam expander, an aperture, a focusing lens, a sample to be tested, and a camera arranged sequentially along the optical path. The laser passes through the beam expander, the aperture, and the focusing lens in sequence to generate an illumination probe that illuminates the sample to be tested on the motion platform. The device also includes a processor, which is used to execute the stacked diffraction position correction imaging method according to any one of claims 1-4.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the stacked diffraction position correction imaging method as described in any one of claims 1 to 4.

Images