A vortex light encoding-based lensless stack imaging method

Abstract

This invention discloses a lensless stacked imaging method based on vortex light coding, which solves the problems of poor system stability, the need for additional image registration preprocessing, low reconstruction efficiency, and limited applicability caused by mechanical scanning in the prior art. It realizes the encoding acquisition and high-resolution reconstruction of target information without mechanical scanning and image registration. The method includes: using the adjustable parameters of the vortex beam to generate a speckle pattern corresponding to the adjustable parameters through a scattering medium, and using the speckle pattern as an illumination probe; when no target is placed in the imaging optical path, calibrating each generated illumination probe to obtain known speckle illumination probes; when the target is placed in the imaging optical path, illuminating the target sequentially with the known speckle illumination probes, and acquiring multiple corresponding low-resolution intensity images by a detector; and reconstructing the high-resolution complex amplitude distribution of the target through an iterative phase retrieval algorithm based on the known speckle illumination probes and multiple low-resolution intensity images.

Description

Technical Field

[0001] This invention relates to the field of optical imaging technology, and in particular to a lensless stacked imaging method based on vortex optical coding. Background Technology

[0002] Optical imaging systems expand human visual perception by visualizing light information, and image recording tools composed of various optical imaging systems have become deeply integrated into daily life. Common optical imaging systems include microscopic systems, photographic systems, and telescopic systems. Lensless stacked microscopy, as a derivative technology of microscopy in the visible light range, removes traditional imaging lenses, illuminates the sample from different angles, or performs a two-dimensional scan of the sample. The light waves carrying the target information diffract over a certain distance, forming a series of low-resolution images on the imaging plane. This technology can achieve diffraction-limited high-resolution imaging while significantly simplifying the optical structure, expanding the field of view, and reducing system costs. Therefore, it has broad application prospects in optical microscopy, medical imaging, and space imaging.

[0003] However, the existence of the scanning mechanism in the experiment limits the stability of the experimental system to some extent. Typically, the scanning process relies on a high-precision control stage, which not only significantly increases experimental cost and complexity but also introduces scanning displacement errors, making it difficult for the system to maintain stability when observing dynamic targets or in extreme environments. Simultaneously, this hardware-level deficiency indirectly affects the quality of acquired images, leading to complex preprocessing steps in the experiment. Scanning errors cause mismatches in image sequence pairing, affecting the reconstruction of subsequent images.

[0004] Therefore, how to avoid the impact of errors caused by mechanical scanning on experimental results, so as to still achieve high-resolution reconstruction of the target image, has become one of the problems that need to be solved in this field.

[0005] By using electronic devices such as LED arrays or spatial light modulators, the lighting angle can be directly controlled and the lighting mode can be quickly programmed and switched without moving the components in the system. This fundamentally eliminates the operation steps of mechanical control scanning, enables rapid switching, and greatly improves the stability of the system. It is currently the most widely used solution and can be flexibly applied to multiple scenarios.

[0006] Multiplexing methods encode target information by specifying wavelength, illumination angle, or spatial pattern, compressing the target information that would otherwise require multiple scans into a single frame or a few frames. By utilizing compressed sensing and joint inversion algorithms, information decoupling and high-resolution reconstruction are achieved, fundamentally eliminating the reliance on mechanical scanning and repeated acquisition. Theoretically, this method can achieve near-limit imaging speeds, meeting the demands of imaging dynamic processes.

[0007] This method unifies the two stages of image registration and spatial stitching into a single joint optimization problem. During iterative reconstruction, the relative displacement between images is treated as a variable and updated iteratively along with the sample complex amplitude. Instead of relying on cross-correlation algorithms, this method utilizes the information redundancy of the acquired images to correct scanning errors during reconstruction, thereby significantly improving system stability and robustness to mechanical errors.

[0008] Traditional lensless stacked microscopy typically requires a high-precision mechanical displacement console to control the light source to illuminate the sample from different angles or to perform two-dimensional scanning. This not only significantly increases system cost and complexity but also introduces displacement errors during continuous scanning, fundamentally limiting the performance and stability of the experimental setup. The impact of mechanical scanning on lensless stacked microscopy experiments is mainly reflected in three aspects: imaging accuracy, system stability, and imaging efficiency. On the one hand, the scanning process generates a certain degree of positioning error, backlash error, and nonlinear motion error, which directly leads to a mismatch between the illumination angle or spatial position and the theoretical imaging model. This results in spatial misalignment of the image, causing reconstruction artifacts and reduced resolution. On the other hand, mechanical scanning is a serial acquisition method, requiring the entire system to stabilize before each image acquisition, greatly limiting the data acquisition speed. This makes lensless stacked microscopy difficult to use in imaging dynamic targets and in extreme environments, thus restricting its application scope.

[0009] Meanwhile, when the displacement distance from the mechanical scan is projected onto the detector, the actual displacement distance changes after being projected onto the detector pixels due to image discretization sampling. This leads to a change in the relative displacement between images after detector sampling, necessitating additional image preprocessing. To obtain the relative displacement between sampled images, registration algorithms such as cross-correlation must be used. However, the registration results heavily depend on image features, making them prone to registration errors in multimodal image features and accumulating errors. Essentially, this reduces a priori problem of computational imaging to an uncertain image processing challenge that must be addressed before reconstruction even begins, and the long scanning time may cause changes in sample state, further compromising data consistency. Summary of the Invention

[0010] This invention provides a lensless stacked imaging method based on vortex optical coding, which solves the problems of poor system stability, the need for additional image registration and preprocessing, low reconstruction efficiency and limited applicability caused by mechanical scanning in the prior art. It realizes the encoding acquisition and high-resolution reconstruction of target information without mechanical scanning and image registration, effectively improving the stability of the system device and the image reconstruction rate and quality.

[0011] This invention provides a lensless stacked imaging method based on vortex light coding, the method comprising: Using the adjustable parameters of the vortex beam, a speckle pattern corresponding to the adjustable parameters is generated through a scattering medium, and the speckle pattern is used as an illumination probe. When no target is placed in the imaging optical path, each generated illumination probe is calibrated to obtain a known speckle illumination probe; When the target is placed in the imaging optical path, the target is illuminated sequentially with a known speckle illumination probe, and the detector acquires multiple corresponding low-resolution intensity images. Based on the known speckle illumination probe and the multiple low-resolution intensity images, the high-resolution complex amplitude distribution of the target is reconstructed using an iterative phase retrieval algorithm.

[0012] In one possible implementation, generating a speckle pattern corresponding to the adjustable parameters of the vortex beam through a scattering medium includes: By loading vortex beams with different topological charges and radial wave vectors onto a spatial light modulator, the incident light is modulated, and after Fourier transform by a lens, a perfect optical vortex beam is generated on the spectral plane. ; The perfect optical vortex beam The light is scattered into the scattering medium, generating a speckle pattern corresponding to the topological charge number and the radial wave vector.

[0013] In one possible implementation, calibrating each generated illumination probe to obtain a known speckle illumination probe when no target is placed in the imaging optical path includes: Without placing a target, the generated speckle patterns are sequentially illuminated onto the detector, which then collects and records the corresponding speckle intensity distribution as a known speckle illumination probe.

[0014] In one possible implementation, when the target is placed in the imaging optical path, the target is sequentially illuminated with a known speckle illumination probe, and a detector acquires multiple corresponding low-resolution intensity images, including: The detector acquires multiple low-resolution intensity images using a forward imaging model; wherein the forward imaging model is represented as: ; in, This represents the complex amplitude of the target; Indicates the first A known speckle illumination probe; Indicates the propagation distance; Represents the diffusion function; Indicates the first A low-resolution intensity image; Indicates the distance propagated through free space The corresponding point spread function; This represents the convolution operation.

[0015] In one possible implementation, the reconstructing of the high-resolution complex amplitude distribution of the target using an iterative phase retrieval algorithm based on the known speckle illumination probe and the plurality of low-resolution intensity images includes: According to the forward imaging model Initialize the preset target complex amplitude ; The following update operations are performed sequentially on multiple low-resolution intensity images, and the preset target complex amplitude is updated after each low-resolution intensity image is processed. The specific update operations include: Each known speckle illumination probe is respectively compared with the preset target complex amplitude. By performing point-by-point multiplication, the outgoing wavefront image departing from the target is obtained. The emitted wavefront image is then processed according to the propagation distance. The wavefront image is propagated to the image sensor plane, resulting in the outgoing wavefront image of the sensor plane. ; The wavefront image emitted from the sensor plane Perform downsampling and analyze the measured intensity image corresponding to the target. The downsampled sensor plane emitted wavefront image The image is updated to obtain the wavefront image of the image sensor. ; The wavefront image of the image sensor will be updated. The image is transmitted back to the plane where the target is located to obtain an updated outgoing wavefront image of the target. ; Based on the updated wavefront image of the target The rPIE algorithm is used to determine the preset target complex amplitude. The preset target complex amplitude is updated to obtain the updated version. ; Until the preset target complex amplitude Convergence occurs, iteration ends, and the reconstructed target complex amplitude distribution is obtained. ; The reconstructed complex amplitude distribution of the target is processed by quasi-focusing based on the angular spectrum diffraction propagation theory to obtain the high-resolution complex amplitude distribution of the reconstructed target.

[0016] In one possible implementation, the preset target complex amplitude Represented as: ; in, Indicates the propagation distance; Represents the diffusion function; Indicates the number of low-resolution intensity images; Indicates the first A low-resolution intensity image; express The nearest neighbor upsampling operation is performed twice.

[0017] In one possible implementation, the sensor plane emits a wavefront image. Represented as: ; in, Represents the diffusion function; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target.

[0018] In one possible implementation, the updated preset target complex amplitude Represented as: ; in, This represents the updated preset target complex amplitude obtained from the previous iteration; Indicates the first A known speckle illumination probe; Indicates the first Updated outgoing wavefront image of the target corresponding to a known speckle illumination probe; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target; These represent the parameters of the rPIE algorithm; This indicates a conjugate operation.

[0019] In one possible implementation, the updated wavefront image of the target Represented as: ; in, This indicates an update to the wavefront image of the image sensor; Represents the diffusion function; Indicates the distance of propagation.

[0020] One or more technical solutions provided in this invention have at least the following technical effects or advantages: This invention utilizes adjustable parameters of a vortex beam to generate a speckle pattern corresponding to these adjustable parameters through a scattering medium. This speckle pattern serves as an illumination probe, allowing for flexible control of the speckle particle size and spatial distribution by electrically adjusting the topological charge and radial wave vector of the vortex beam. This generates a series of independent, structured illumination probes, avoiding the use of mechanical scanning devices and improving system stability and flexibility. When no target is placed in the imaging optical path, the generated illumination probes are calibrated to obtain known speckle illumination probes. This ensures that the illumination probe information is completely known during subsequent imaging, eliminating the need for online estimation, simplifying the reconstruction algorithm, and improving reconstruction accuracy and efficiency. Furthermore, the calibration process requires no target and is easy to operate. When a target is placed in the imaging optical path, the... The target is illuminated using a known speckle illumination probe, and multiple low-resolution intensity images are acquired by the detector. The target information is spatially encoded using the known speckle probe, and the high-frequency components of the target are moved into the detector's passband, allowing the low-resolution images to carry high-resolution information. The acquisition process does not require mechanical scanning or image registration, improving imaging speed and system robustness. Based on the known speckle illumination probe and the multiple low-resolution intensity images, the high-resolution complex amplitude distribution of the target is reconstructed using an iterative phase retrieval algorithm. Using the known probe as prior information, the target is accurately restored from the intensity images through phase retrieval, giving full play to the advantages of information redundancy, correcting system errors, and ultimately achieving high-resolution lensless stacked imaging without mechanical scanning. Attached Figure Description

[0021] Figure 1 A flowchart of the lensless stacked imaging method based on vortex optical coding provided in this embodiment of the invention; Figure 2 These are speckle patterns under different modulation conditions provided in embodiments of the present invention; Figure 3 The original image acquired during the experiment, i.e. the reconstructed result image, is provided for the embodiments of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0023] This invention provides a lensless stacked imaging method based on vortex optical coding, see [link to relevant documentation]. Figure 1 The method includes the following steps S101 to S104.

[0024] S101 utilizes the adjustable parameters of the vortex beam to generate a speckle pattern corresponding to the adjustable parameters through a scattering medium, and uses the speckle pattern as an illumination probe. Specifically, in step S101, a speckle pattern corresponding to the adjustable parameters is generated through a scattering medium using the adjustable parameters of the vortex beam, including the following steps S1011 to S1012.

[0025] S1011 loads vortex beams with different topological charges and radial wave vectors onto a spatial light modulator to modulate the incident light. After Fourier transform by a lens, a perfect optical vortex beam is generated on the spectral plane. ; S1012 will produce a perfect optical vortex beam. The light is scattered into the scattering medium, generating a speckle pattern corresponding to the topological charge number and radial wave vector.

[0026] For example, the illumination source used in this invention is a laser source, which has limited effect on speckle pattern modulation. Based on traditional methods, a vortex illumination pattern is introduced, and its two special properties—topological charge number—are adjusted. and radial wave vector The structure of the light spot illuminating the scattering medium is controlled to change the particle size of the speckle pattern, thereby using the different speckle patterns formed by the control to encode the target information.

[0027] By loading optical field patterns with different topological charges and radial wave vectors onto a spatial light modulator, additional radial oscillation terms and vortex phase terms will be added to the original incident light field: (1.1) in, The light field incident on the spatial light modulator, It is a radial oscillation term. This is the vortex phase term.

[0028] After the beam generated by the spatial light modulator is passed through the lens and subjected to Fourier transform, a perfect optical vortex beam (POV) is generated on the spectral plane.

[0029] (1.2) in, Let be the radius of the halo on the spectral plane. For frequency.

[0030] S102, when no target is placed in the imaging optical path, calibrate each generated illumination probe to obtain known speckle illumination probes; Specifically, in step S102, when no target is placed in the imaging optical path, each generated illumination probe is calibrated to obtain known speckle illumination probes, including: Without placing a target, the generated speckle patterns are sequentially illuminated onto the detector, which then collects and records the corresponding speckle intensity distribution as a known speckle illumination probe.

[0031] For example, when the aforementioned perfect optical vortex beam (POV) is irradiated onto a scattering medium (frosted glass), the resulting illumination probe undergoes a significant change. This can be seen from the above equation. Proportional to Furthermore, since the speckle size is inversely proportional to the effective illumination spot diameter, the radial wave vector can be increased. To increase the effective illumination spot diameter and thus reduce the halo radius This results in smaller and denser speckle particles. Simultaneously, by changing the topological charge... Additional phase modulation can be introduced into the scattering pattern, changing the interference mode of the speckle field and making the speckle particles more compact.

[0032] S103, when the target is placed in the imaging optical path, the target is illuminated sequentially with a known speckle illumination probe, and the detector acquires multiple corresponding low-resolution intensity images; Specifically, in step S103, when the target is placed in the imaging optical path, the target is sequentially illuminated with a known speckle illumination probe, and the detector acquires multiple corresponding low-resolution intensity images, including: The detector acquires multiple low-resolution intensity images using a forward imaging model; the forward imaging model is represented as follows: (1.3) in, Represents the complex amplitude of the target; Indicates the first A known speckle illumination probe; Indicates the propagation distance; Represents the diffusion function; Indicates the first A low-resolution intensity image; Indicates the distance propagated through free space The corresponding point spread function; This represents the convolution operation.

[0033] For example, based on the aforementioned vortex beam generation mechanism, this invention utilizes a spatial light modulator to load a series of vortex beam patterns. The emitted light waves, after passing through a scattering medium (frosted glass), form a special speckle pattern. First, a detector is used to acquire and calibrate speckle patterns modulated by different vortex beams. Then, the target is added to the imaging optical path, and a series of low-resolution images containing the target under illumination with different speckle patterns are acquired. In the illumination optical path, the non-uniform speckle generated after the coherent light passes through the scattering medium typically leads to a decrease in image resolution. However, from a spectral perspective, the target and the speckle formed after passing through the scattering medium can, to some extent, shift the high-frequency components of the target into the system's pupil function. Therefore, the speckle pattern can be considered as an illumination spot with a certain spatial distribution. The low-resolution image captured by the system's image plane contains more high-frequency components of the target, which is crucial for subsequent high-resolution target reconstruction. The forward imaging model of this invention can be expressed as formula (1.3).

[0034] S104, based on known speckle illumination probes and multiple low-resolution intensity images, reconstructs the high-resolution complex amplitude distribution of the target using an iterative phase retrieval algorithm.

[0035] Specifically, in step S104, based on the known speckle illumination probe and multiple low-resolution intensity images, the high-resolution complex amplitude distribution of the target is reconstructed using an iterative phase retrieval algorithm, including: (1) Based on the forward imaging model Initialize the preset target complex amplitude Here, the target complex amplitude is preset. Represented as: (1.4) in, Indicates the propagation distance; Represents the diffusion function; Indicates the number of low-resolution intensity images; Indicates the first A low-resolution intensity image; express The nearest neighbor upsampling operation is performed twice.

[0036] (2) Perform the following update operations sequentially on multiple low-resolution intensity images, and update the preset target complex amplitude after processing each low-resolution intensity image. The specific update operations include: (2.1) Connect each known speckle illumination probe to the preset target complex amplitude respectively. By performing point-by-point multiplication, the outgoing wavefront image leaving the target is obtained. And based on the propagation distance, the outgoing wavefront image The wavefront image is propagated to the image sensor plane, resulting in the outgoing wavefront image of the sensor plane. Here, the sensor plane outputs a wavefront image. Represented as: (1.5) in, Represents the diffusion function; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target.

[0037] (2.2) Wavefront image emitted from the sensor plane Perform downsampling and analyze the measured intensity image corresponding to the target. The downsampled sensor plane emitted wavefront image The image is updated to obtain the wavefront image of the image sensor. ; (2.3) Update the wavefront image of the image sensor. The image is transmitted back to the plane where the target is located, resulting in an updated image of the target's outgoing wavefront. ; Here is the updated wavefront image of the target. Represented as: (1.6) in, This indicates an update to the wavefront image of the image sensor; Represents the diffusion function; Indicates the distance of propagation.

[0038] (3) Until the preset target complex amplitude is reached Convergence occurs, iteration ends, and the reconstructed target complex amplitude distribution is obtained. ; Here, the updated preset target complex amplitude Represented as: (1.7) in, This represents the updated preset target complex amplitude obtained from the previous iteration; Indicates the first A known speckle illumination probe; Indicates the first Updated outgoing wavefront image of the target corresponding to a known speckle illumination probe; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target; This represents the parameter of the rPIE algorithm, which is usually set to 1; This indicates a conjugate operation.

[0039] (4) Based on the angular spectrum diffraction propagation theory, the reconstructed target complex amplitude distribution is quasi-focused to obtain the high-resolution complex amplitude distribution of the reconstructed target.

[0040] For example, the corresponding target reconstruction process is as follows: Step 1: Initialize the estimated preset target complex amplitude For the specific formula, please refer to formula (1.4).

[0041] Step 2: Apply the speckled pattern With the preset target complex amplitude The outgoing wavefront image departing from the target is obtained by pixel-by-pixel product. The outgoing wavefront image will then be displayed. spread Obtain the wavefront image of the image sensor plane. Specifically, the image of the departing wavefront leaving the target. Represented as: (1.8) Sensor planar emitted wavefront image See Formula (1.5) for details.

[0042] Step 3: Image the wavefront emitted from the sensor plane on the image sensor plane. Perform downsampling and utilize the measured intensity image corresponding to the target. Update sensor planar outgoing wavefront image Specifically, the wavefront image emitted from the sensor plane on the image sensor plane. The result of performing the downsampling operation is expressed as follows: (1.9) Update the wavefront image of the image sensor Represented as: (1.10) in, This represents the floor function. express A downsampling operation of 10 times.

[0043] Step 4: Update the wavefront image of the image sensor. Propagate back to the plane where the target is located, see formula (1.6) for details.

[0044] Step 5: Update the target complex amplitude using the rPIE algorithm The updated preset target complex amplitude This is represented by formula (1.7).

[0045] Step 6: For the results reconstructed by the above algorithm Because the distance between the target and the camera cannot be precisely measured in the experiment, the recovered image is out of focus. Therefore, it is necessary to apply the angular spectrum diffraction propagation theory to... Perform focusing. Here, it's important to note that... and All represent the final updated target complex amplitude. The z-coordinate is included here because the focusing process needs to be propagated back along the optical axis (z-axis). The z-coordinate is not labeled because the entire process takes place on the (x,y) plane. Can be written as .

[0046] (1.11) (1.12) (1.13) (1.14) in, This represents the spatial frequency of the light field function in the x-direction. ; This represents the spatial frequency of the light field function in the y-direction. ; Indicates the distance traveled along the optical axis; This represents the projection of wave vector k onto the z-direction. ; Represents the wave vector, characterizing the direction of light field propagation. ; express The spatial spectrum distribution at a distance z from the origin on the z-axis can be written as: ; The angular spectrum propagation function is... The Fourier transform of a light field function is obtained by multiplying it by the function if the light field function has traveled a distance z along the z-axis. This represents the light field distribution after the light field function has propagated a distance z.

[0047] When ordinary random speckle patterns pass through a rough surface, if the sample has a certain thickness, the speckle pattern is prone to deformation and degradation as it passes through the sample. This makes the imaging system extremely sensitive to the focal plane. If the target position is slightly off or has a certain thickness, the acquired diffraction pattern will lose contrast, causing the algorithm to fail to accurately restore the phase information and produce severe artifacts.

[0048] The scattering pattern modulated by vortex light mentioned in this invention exhibits a "radial" pattern and properties similar to a non-diffractive beam, thus maintaining its shape within a certain distance along the propagation direction. This helps improve the depth of field in lensless imaging. Even with variations in sample thickness, the speckle pattern retains a relatively fine and stable structure. This enables the system to perform high-quality reconstruction of complex and thick imaging targets while maintaining stable imaging resolution, demonstrating its applicability to a variety of complex imaging targets.

[0049] This invention can directly adjust the size of speckle particles by electronically controlling the topological charge and radial wave vector of a vortex beam, without requiring the movement or control of other optical components. Therefore, it can serve as a complete system device, capable of achieving optimal reconstruction results more quickly by selecting different parameters for imaging different targets, and has broad application prospects.

[0050] In a simulation experiment provided by this invention, based on Figure 2 The different modulated speckle patterns shown and Figure 3 The experimental objectives are shown below. Simulation experiments are conducted to verify the effectiveness of the method of the present invention: 1. Speckle pattern generation. According to the method of claim 2, by adjusting the topological charge of the vortex beam... and radial wave vector , generation and Figure 2 The three corresponding typical speckle patterns are as follows: Figure 2 (a) Unmodulated speckle pattern The speckle particle size is 35.69 mm. . Figure 2 (b) is for modulating speckle patterns ( ), speckle particle size . Figure 2 (c) is for modulating speckle patterns ( ), speckle particle size .

[0051] 2. Imaging and Acquisition Simulation. First, the above 30 sets of speckle patterns were calibrated without a target to obtain known speckle illumination. Then with Figure 3 (a) shows the USAF resolution test board as the target, and sequentially uses... The target was illuminated, and 30 low-resolution intensity images were acquired based on a forward imaging model.

[0052] 3. Image reconstruction simulation.

[0053] Initialization: Upsample and average 30 low-resolution images to obtain the initial target complex amplitude; Iterative update: Each image and its corresponding known speckle pattern are used sequentially to perform outgoing wavefront calculation, forward propagation, intensity update, backward propagation and rPIE update. The target complex amplitude is updated after each image is processed. After 50 iterations until convergence, the diagonal spectrum propagation distance is refocused.

[0054] 4. Reconstruction Results. The reconstruction results are as follows: Figure 3 As shown in (b). Compared to the original target Figure 3 Compared to (a), the reconstructed image clearly restores the details of the 3rd line pair in group 9, with no obvious artifacts or misalignments. Compared to unmodulated speckle illumination (using only...) Figure 2 Compared to (a) speckle), the resolution is improved.

[0055] 5. Verification Conclusions. Simulation results verify that the method of the present invention can achieve high-resolution lensless imaging under mechanical scanning conditions by combining speckle illumination modulated by vortex beam with an iterative reconstruction algorithm.

[0056] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this invention can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0057] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A lensless stacked imaging method based on vortex optical coding, characterized in that, include: Using the adjustable parameters of the vortex beam, a speckle pattern corresponding to the adjustable parameters is generated through a scattering medium, and the speckle pattern is used as an illumination probe. When no target is placed in the imaging optical path, each generated illumination probe is calibrated to obtain a known speckle illumination probe; When the target is placed in the imaging optical path, the target is illuminated sequentially with a known speckle illumination probe, and the detector acquires multiple corresponding low-resolution intensity images. Based on the known speckle illumination probe and the multiple low-resolution intensity images, the high-resolution complex amplitude distribution of the target is reconstructed using an iterative phase retrieval algorithm.

2. The lensless stacked imaging method based on vortex optical coding according to claim 1, characterized in that, The method of generating a speckle pattern corresponding to the adjustable parameters of a vortex beam through a scattering medium includes: By loading vortex beams with different topological charges and radial wave vectors onto a spatial light modulator, the incident light is modulated, and after Fourier transform by a lens, a perfect optical vortex beam is generated on the spectral plane. ; The perfect optical vortex beam The light is scattered into the scattering medium, generating a speckle pattern corresponding to the topological charge number and the radial wave vector.

3. The lensless stacked imaging method based on vortex optical coding according to claim 1, characterized in that, The process of calibrating each generated illumination probe when no target is placed in the imaging optical path to obtain known speckle illumination probes includes: Without placing a target, the generated speckle patterns are sequentially illuminated onto the detector, which then collects and records the corresponding speckle intensity distribution as a known speckle illumination probe.

4. The lensless stacked imaging method based on vortex optical coding according to claim 1, characterized in that, When the target is placed in the imaging optical path, the target is sequentially illuminated with a known speckle illumination probe, and a detector acquires multiple corresponding low-resolution intensity images, including: The detector acquires multiple low-resolution intensity images using a forward imaging model; wherein the forward imaging model is represented as: ； in, This represents the complex amplitude of the target; Indicates the first A known speckle illumination probe; Indicates the propagation distance; Represents the diffusion function; Indicates the first A low-resolution intensity image; Indicates the distance propagated through free space The corresponding point spread function; This represents the convolution operation.

5. The lensless stacked imaging method based on vortex optical coding according to claim 1, characterized in that, The process of reconstructing the high-resolution complex amplitude distribution of the target using an iterative phase retrieval algorithm based on the known speckle illumination probe and the multiple low-resolution intensity images includes: According to the forward imaging model Initialize the preset target complex amplitude ; The following update operations are performed sequentially on multiple low-resolution intensity images, and the preset target complex amplitude is updated after each low-resolution intensity image is processed. The specific update operations include: Each known speckle illumination probe is respectively compared with the preset target complex amplitude. By performing point-by-point multiplication, the outgoing wavefront image departing from the target is obtained. The emitted wavefront image is then processed according to the propagation distance. The wavefront image is propagated to the image sensor plane, resulting in the outgoing wavefront image of the sensor plane. ; The wavefront image emitted from the sensor plane Perform downsampling and analyze the measured intensity image corresponding to the target. The downsampled sensor plane emitted wavefront image The image is updated to obtain the wavefront image of the image sensor. ; The wavefront image of the image sensor will be updated. The image is transmitted back to the plane where the target is located to obtain an updated outgoing wavefront image of the target. ; Based on the updated wavefront image of the target The rPIE algorithm is used to determine the preset target complex amplitude. The preset target complex amplitude is updated to obtain the updated version. ; Until the preset target complex amplitude Convergence occurs, iteration ends, and the reconstructed target complex amplitude distribution is obtained. ; The reconstructed complex amplitude distribution of the target is processed by quasi-focusing based on the angular spectrum diffraction propagation theory to obtain the high-resolution complex amplitude distribution of the reconstructed target.

6. The lensless stacked imaging method based on vortex optical coding according to claim 5, characterized in that, The preset target complex amplitude Represented as: ； in, Indicates the propagation distance; Represents the diffusion function; Indicates the number of low-resolution intensity images; Indicates the first A low-resolution intensity image; express The nearest neighbor upsampling operation is performed twice.

7. The lensless stacked imaging method based on vortex optical coding according to claim 5, characterized in that, The sensor plane emits a wavefront image. Represented as: ； in, Represents the diffusion function; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target.

8. The lensless stacked imaging method based on vortex optical coding according to claim 5, characterized in that, The updated preset target complex amplitude Represented as: ； in, This represents the updated preset target complex amplitude obtained from the previous iteration; Indicates the first A known speckle illumination probe; Indicates the first Updated outgoing wavefront image of the target corresponding to a known speckle illumination probe; Indicates the first The outgoing wavefront image of a known speckle illumination probe corresponding to the target; These represent the parameters of the rPIE algorithm; This indicates a conjugate operation.

9. The lensless stacked imaging method based on vortex optical coding according to claim 5, characterized in that, The updated wavefront image of the target Represented as: ； in, This indicates an update to the wavefront image of the image sensor; Represents the diffusion function; Indicates the distance of propagation.

Images