Multi-depth holographic method based on random noise multiple orthogonality

By introducing random noise multiple orthogonality and time multiplexing methods in multi-deep holographic technology, crosstalk and speckle problems are solved, image quality and signal-to-noise ratio are improved, and fast and high-quality multi-deep holographic reconstruction is achieved.

CN118963091APending Publication Date: 2024-11-15SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411233082.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

There is a crosstalk problem in the existing multi-depth holographic technology, resulting in low quality of reconstruction images and limited information capacity. The optimization algorithm is complex in the calculation and random speckle in the generated images, which reduces the signal-to-noise ratio.

Method used

Using a multi-depth holographic method based on the orthogonality of random noise multiples, by generating random noise seeds superimposing random noise of different multiples on different depth target images, two-dimensional Dirac comb function sampling and inverse angle spectroscopy operation, combined with time complex Methods are used to quickly play multiple sets of holograms on a spatial light modulator to observe the reconstructed images after time integration.

Benefits of technology

It effectively suppresses crosstalk between various depth planes, improves the computing speed, imaging quality and depth resolution of multi-depth holographic zombies, and significantly improves the signal-to-noise ratio.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963091A_ABST
    Figure CN118963091A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-depth holographic method based on random noise multiple orthogonality. According to the method, the orthogonality between random noise multiples is utilized, pre-sampling is carried out after random noise with different multiples corresponding to the depth number of a target image is superposed on the target image, and complex amplitude holography of images with different depths is calculated and solved by utilizing a reverse angular spectrum method; and summing the complex amplitude holograms of different depths to obtain a multi-depth hologram. In order to further improve the signal-to-noise ratio of the multi-depth hologram based on the random noise multiple orthogonality, the method is also combined with a time multiplexing method, namely, the holograms with the same target image but different additional uniform random noises are quickly and circularly played, and the reconstructed image after time integration is observed. According to the method, the orthogonality between random noise multiples is utilized, and the signal-to-noise ratio of the reconstructed image is improved by combining time multiplexing, so that multi-depth holography with high capacity, low crosstalk and high signal-to-noise ratio is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the fields of light field regulation, holographic display, and holographic storage and encryption, and in particular to a multi-depth holographic method based on the orthogonality of random noise multiples. Background Art

[0002] Holography can reconstruct properties such as the amplitude and phase of a light field, and has been applied in the fields of three-dimensional display, light field control, optical encryption, and microscopic imaging. With the development of electronic computers and wavefront shaping technology, holography is no longer limited to single-plane holography, but has gradually developed towards multi-depth holography to achieve true three-dimensional display and three-dimensional light field reconstruction. However, since the propagation of electromagnetic fields needs to satisfy Maxwell's equations, a light field at a certain depth will produce diffraction intensity at another depth when it is diffracted to another depth, causing crosstalk to inevitably appear in multi-depth holography. In order to reduce the crosstalk between multi-depth planes, researchers have applied a variety of optimization algorithms in multi-depth holography, including the improved Gerchberg-Saxton (GS) algorithm and the stochastic gradient descent algorithm, which can reduce the crosstalk between planes of multi-depth holography to a certain extent. However, due to the high computational complexity of the optimization algorithm, the multi-depth hologram generated by the optimization algorithm still has residual crosstalk in its reconstructed image, which will also limit the information capacity of the multi-depth image. Therefore, it is necessary to develop a method that can quickly and high-quality realize multi-depth holography.

[0003] In order to improve the reconstruction quality of multi-depth holography, researchers proposed the concept of applying noise engineering, that is, introducing non-correlated random phases in the generation of holograms, so that the crosstalk intensity at different depths is diffused in space, thereby reducing the intensity of crosstalk relative to the multi-depth target image within the observation range. However, since this method introduces random phases in the process of holographic generation, a large number of random specks will appear in the reconstructed image, thereby reducing the signal-to-noise ratio of the reconstructed image. Therefore, it is necessary to propose a multi-depth holographic method with short time consumption, high imaging quality and high depth resolution to promote the development of holographic display, holographic storage and encryption technologies. Summary of the invention

[0004] In order to improve the imaging quality of multi-depth holography, the present invention proposes a multi-depth holography method based on the orthogonality of random noise multiples. In order to achieve the above purpose, the technical solution proposed by the present invention is as follows:

[0005] A multi-depth holographic method based on orthogonality of random noise multiples comprises the following steps:

[0006] Step 1: Generate a random noise seed and superimpose corresponding multiples of random noise on the target image at different depths;

[0007] Step 2: Use the two-dimensional Dirac comb function to sample images at different depths to reduce crosstalk between different pixels;

[0008] Step 3: Use the inverse angular spectrum method to calculate the complex amplitude hologram of images at different depths;

[0009] Step 4: sum the complex amplitude holograms of images at different depths to obtain a multi-depth hologram;

[0010] Step 5: Using different random noise seeds, repeat steps 1-4 to obtain multiple sets of multi-depth holograms;

[0011] Step 6: Rapidly play multiple sets of multi-depth holograms on the spatial light modulator and observe the reconstructed image after time integration.

[0012] Furthermore, in step 2, the sampling interval and sampling point size of the two-dimensional Dirac comb function will affect the quality of the reconstructed image and need to be selected based on merit.

[0013] Compared with the prior art, the technology of the present invention has the following outstanding characteristics and advantages:

[0014] The present invention introduces different multiples of random noise in the process of generating multi-depth holograms, that is, superimposing different multiples of random noise phases on target images at different depths. Since the inner product of different multiples of random noise phases is zero, the orthogonality condition is satisfied, and thus the crosstalk between planes at different depths can be effectively suppressed. In order to further improve the signal-to-noise ratio of multi-depth holography based on the orthogonality of random noise multiples, the present invention also combines a time multiplexing method, that is, quickly looping holograms with the same target image but different uniform random noises, and observing the reconstructed image after time integration. The present invention can effectively improve the computing speed, imaging quality and depth resolution of multi-depth holography. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to further explain the embodiments of the present invention in detail, the following briefly introduces the drawings used in the embodiments. It is necessary to point out that the drawings are only some embodiments of the present invention and do not mean any limitation on the protection scope of the present invention.

[0016] Figure 1 It is an implementation flow chart of the present invention.

[0017] Figure 2 Schematic diagram of a multi-depth holographic method based on orthogonality of random noise multiples in Example 1 of the present invention. Figure 2 (a) is the hologram encoding process (first frame), Figure 2 (b) is a schematic diagram of the time multiplexing process.

[0018] Figure 3This is a schematic diagram of the relationship between hologram parameters and reconstructed image quality, taking the image shown in the figure as an example in Example 1 of the present invention. Figure 3 (a) is the intensity distribution diagram of the point-shaped Gaussian comb function on an imaging plane. Figure 3 (b) is a schematic diagram of the crosstalk intensity generated on another imaging surface after the image in (a) propagates a certain distance. Figure 3 (c) is a multi-depth holographic simulation effect diagram. The hologram parameters are set as follows: the total depth is 5, the reconstructed image of the first depth is selected for display, the hologram resolution is 1536×1536, the interval between adjacent planes is 10mm, the sampling spacing of the point Gaussian comb function is 125μm and the Gaussian function beam width is 30μm. Figure 3 (dg) are changed respectively Figure 3 (c) The simulation results obtained after simulating a certain hologram parameter. Figure 3 The hologram resolution of (d) was adjusted to 3072 × 3072; Figure 3 The interval between adjacent planes of (e) was adjusted to 20 mm; Figure 3 The sampling spacing of the point-shaped Gaussian comb function of (f) is adjusted to 175 μm; Figure 3 The Gaussian function beam width of (g) was adjusted to 15 μm.

[0019] Figure 4 It is the result of the multi-depth holographic method based on the orthogonality of random noise multiples in Example 1 of the present invention, taking the image shown in the figure as an example. Figure 4 The distance between each plane is 10 mm. Simulations of 1 frame, 4 frames, 9 frames, 16 frames and 25 frames multiplexing were carried out respectively, and the peak signal-to-noise ratio (PSNR) was used as the image quality evaluation index. DETAILED DESCRIPTION

[0020] The following will be combined with the accompanying drawings in the technical embodiments of the present invention to clearly and completely describe the technical solutions in the technical embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the technology of the present invention, not all of them.

[0021] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0022] Example 1

[0023] Figure 1 It is a flow chart of the present invention, comprising the following steps:

[0024] Step 1: Generate a random noise seed and superimpose corresponding multiples of random noise on the target image at different depths;

[0025] Step 2: Use the two-dimensional Dirac comb function to sample images at different depths to reduce crosstalk between different pixels;

[0026] Step 3: Use the inverse angular spectrum method to calculate the complex amplitude hologram of images at different depths;

[0027] Step 4: sum the complex amplitude holograms of images at different depths to obtain a multi-depth hologram;

[0028] Step 5: Using different random noise seeds, repeat steps 1-4 to obtain multiple sets of multi-depth holograms;

[0029] Step 6: Rapidly play multiple sets of multi-depth holograms on the spatial light modulator and observe the reconstructed image after time integration.

[0030] Furthermore, in step 1, a uniformly distributed random phase is used as a random phase seed, such as Figure 2 As shown in (a), the random phase is exp[i·q·ψ j (x, y)], where q represents the qth depth, j represents the jth frame, and ψ j (x,y) represents the random noise seeds of different frames.

[0031] In the multi-depth holographic method based on the orthogonality of random noise multiples, the holograms corresponding to different frames are multi-depth images of the same target, but the phases of the random noise superimposed on the target image are different. This method achieves the effect of reducing the crosstalk and speckle intensity in the reconstructed image by time-integrating the multi-depth images reconstructed from holograms of different frames.

[0032] In step 2, the sampling and hologram parameters will significantly affect the performance of the multi-depth hologram for the following reasons.

[0033] Theoretically, the orthogonality between random noise multiples can completely eliminate the crosstalk and noise of multi-depth holograms. However, in practice, a point-like Gaussian comb function is needed to sample the target image. At the same time, the resolution of the hologram is also limited, so residual crosstalk will appear in the reconstructed image.

[0034] Furthermore, the residual crosstalk intensity is analyzed. In multi-depth holography, the crosstalk within a certain depth comes from the diffraction intensity of the light field at other depths in the plane. Since the intervals between different depths are short, the Rayleigh-Sommerfeld diffraction integral formula is used for calculation, which can be expressed as:

[0035]

[0036] Wherein, E1 represents the output light field, (x, y) represents the plane coordinates of the output surface, z represents the distance between the input surface and the output surface, k represents the wave number, (ξ, η) represents the plane coordinates of the input surface, and E0 represents the input light field.

[0037] In numerical calculation, it is necessary to discretize equation (1). In order to simplify the analysis process, we first consider the diffraction of a point Gaussian beam between adjacent planes. Figure 3 (a) describes the intensity distribution of the point-shaped Gaussian function on the imaging plane, where the Gaussian point array is regarded as image information. However, when the Gaussian point function is transmitted a short distance to the adjacent plane, the diffraction intensity will become crosstalk on the adjacent plane, such as Figure 3 (b) Therefore, E0 should be a Gaussian point function with a random phase, E1 should be the generated crosstalk complex amplitude, and the discretized equation (1) can be expressed as:

[0038]

[0039] Where Δz represents the plane spacing, (a, b) represents the coordinate subscript number of the output surface, Δx0Δy0 represents the image pixel area, w represents the beam width of the point Gaussian beam, and ψ represents the random phase superimposed on the Gaussian function.

[0040] Therefore, the crosstalk intensity can be expressed as:

[0041]

[0042] The crosstalk intensity is analyzed using a statistical method, that is, the mathematical expectation and variance of the crosstalk intensity are calculated. After a series of derivations, the expression of the mathematical expectation of the crosstalk intensity is obtained:

[0043]

[0044] In the formula, integration is used to approximate the summation process.

[0045] Using a similar method to calculate the variance of the crosstalk intensity, we can obtain:

[0046]

[0047] D[I crosstalk (x,y)]=E{γI crosstalk (x,y)] 2}-E 2 [I crosstalk (x,y)]=E 2 [I crosstalk (x,y)] (6)

[0048] Assuming Δz>>w and x,y>>w, the mean and variance of the crosstalk intensity can be approximately expressed as:

[0049]

[0050]

[0051] In the formula, ρ 2 =x 2 +y 2 .

[0052] Equations (7) and (8) show that the mean and variance of the crosstalk intensity decrease with the improvement of resolution (i.e., the decrease of Δx0Δy0), the increase of the plane spacing, and the decrease of the beam width of the point-shaped Gaussian beam. In addition, as the sampling spacing of the point-shaped Gaussian comb function increases, the spacing between different sampling points becomes larger, and the total crosstalk intensity will also decrease. Figure 3 (cg) shows the simulation results after changing the hologram parameters, which verifies the correctness and effectiveness of the proposed method. However, it is necessary to pay attention to the negative effects that may be caused by the change of hologram parameters. For example, the increase of hologram resolution will lead to the increase of equipment complexity, and the larger sampling spacing will lead to the reduction of spatial frequency of reconstructed multi-depth images. Therefore, the relevant parameters need to be selected according to the specific situation.

[0053] In step 3, the complex amplitude distribution calculated by the inverse angular spectrum method is recorded as Ω q,j (ξ,η).

[0054] In step 4, the calculated multi-depth hologram in a certain frame can be expressed as:

[0055]

[0056] In step 5, the effectiveness of time multiplexing will be further verified. In steps 1-4, the imaging quality of multi-depth holography is significantly improved by utilizing the orthogonality between multiple random noises. However, since this method introduces random variables in the hologram generation process, obvious noise intensity will appear in the target image. Therefore, it is necessary to apply time multiplexing to enhance the signal-to-noise ratio of the reconstructed multi-depth image.

[0057] It can be seen from this step that the actual operation of time multiplexing is to quickly switch and play multiple groups of multi-depth holograms on the spatial light modulator. Assuming that the total number of time multiplexing groups is r, since the random phases superimposed on the target images with different frame numbers are independent and uniformly statistically distributed, the time integration in this method is equivalent to averaging the intensity of the reconstructed images with different frame numbers. According to the probability theory, in the reconstructed image after time integration, the mean of the crosstalk intensity remains unchanged, while its variance is reduced to 1 / r of the original crosstalk intensity variance. Taking equations (7) and (8) as examples, after time integration, the mean and variance of the crosstalk intensity can be expressed as

[0058]

[0059]

[0060] It can be seen that time multiplexing will significantly reduce the variance of the crosstalk (noise) intensity in the reconstructed image and effectively improve the image signal-to-noise ratio.

[0061] In step 6, the dynamic switching frequency of the SLM needs to be much greater than the exposure time of the observation camera to achieve the purpose of observing the time average of the intensity of the reconstructed image. Figure 4 As shown, the hologram resolution is 1024×1024.

[0062] The interval between adjacent planes is 10 mm, the sampling interval of the point-shaped Gaussian comb function is 125 μm, and the Gaussian point beam width is 30 μm. The simulations were performed with the total number of time multiplexing frames being 1, 4, 9, 16, and 25. The simulation results are consistent with the theoretical analysis, that is, with the increase of the total number of time multiplexing groups r, the peak signal-to-noise ratio of the reconstructed image increases, and the image quality is significantly improved.

[0063] In summary, Example 1 comprehensively describes the multi-depth holographic method assisted by time multiplexing based on the orthogonality of random phase multiples, and explains the specific encoding process and observation process of the present invention. At the same time, Example 1 demonstrates the influence of different hologram parameters on the reconstruction quality, and makes suggestions on the selection of hologram parameters. Furthermore, Example 1 verifies the effectiveness of the time multiplexing method and demonstrates a high multi-depth image reconstruction quality.

[0064] The specific description of the above embodiments provides further detailed description of the purpose, technical solution and beneficial effects of the invention. The above specific implementation methods are only used to illustrate the technical solution of the present invention and cannot be used to limit the scope of rights. It should be particularly pointed out that for those skilled in the art, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A multi-depth holographic method based on orthogonality of random noise multiples, characterized in that: The following steps are involved: Step 1: Generate a random noise seed and superimpose corresponding multiples of random noise on the target image at different depths; Step 2: Use the two-dimensional Dirac comb function to sample images at different depths to reduce crosstalk between different pixels; Step 3: Use the inverse angular spectrum method to calculate the complex amplitude hologram of images at different depths; Step 4: sum the complex amplitude holograms of images at different depths to obtain a multi-depth hologram; Step 5: Using different random noise seeds, repeat steps 1-4 to obtain multiple sets of multi-depth holograms; Step 6: Rapidly play multiple sets of multi-depth holograms on the spatial light modulator and observe the reconstructed image after time integration.

2. The multi-depth holographic method based on orthogonality of random noise multiples according to claim 1, characterized in that: In step 2, the sampling interval and sampling point size of the two-dimensional Dirac comb function will affect the quality of the reconstructed image and need to be selected on an optimal basis.