Multi-focal-plane hologram generation and optimization method based on ACGS algorithm

Abstract

The invention discloses a multi-focal-plane hologram generation and optimization method based on an ACGS algorithm, and the method comprises the steps: setting the phase distribution and amplitude of a hologram generated by a spatial light modulator according to target image information, and generating an initial multi-focal-plane hologram; performing a simulation experiment on the initial multi-focal-plane hologram, and testing and evaluating key indexes of an optical path system and the multi-focal-plane hologram to optimize a holographic algorithm of the multi-focal-plane hologram so as to form the multi-focal-plane hologram with aberration compensation; further optimizing the phase distribution of the multi-focal plane hologram by using an iterative algorithm so as to improve the multi-focal plane imaging quality; when the number of iterations reaches a preset value, circulation is stopped, and final phase information and a holographic algorithm are introduced into the multi-focal-plane hologram; according to the method, multiple focal planes can be reconstructed at the same time, and the weight is dynamically adjusted according to the image error and the local contrast, so that the reconstruction speed is improved, information loss is reduced, and the image quality is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-focal plane holograms, and in particular to a multi-focal plane hologram generation and optimization method based on an ACGS algorithm. Background Art

[0002] Currently, most light field screens include an image generating unit, a window unit and an image magnifying unit, wherein the image generating unit is used to emit image light to the window unit, the window unit reflects the image light from the image generating unit to the image magnifying unit, the image magnifying unit is used to reflect the image light from the window unit to the window unit, and the window unit transmits the image light from the image magnifying unit to the human eye position.

[0003] The iterative process of the traditional GS algorithm first takes the amplitude distribution of the target image and the random phase of equal magnitude as the initial light field distribution. After Fourier transformation, the complex amplitude distribution in the frequency domain is obtained, and then the frequency domain constraint is applied to retain the phase information in the frequency domain and replace its amplitude with the known light field amplitude. After the inverse Fourier transform, the propagation of the input surface light field is obtained, and the spatial domain constraint is applied. The process is iterated until the convergence condition is met.

[0004] Therefore, the disadvantages of the existing traditional GS algorithm include: slow convergence speed, easy to fall into local optimal solution, strong dependence on initial value, etc. Summary of the invention

[0005] The purpose of the present invention is to provide a multi-focal plane hologram generation and optimization method based on the ACGS algorithm to solve the technical problems of the GS algorithm in the prior art, such as slow convergence speed, easy to fall into local optimal solution, and strong dependence on initial values.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] A multi-focal plane hologram generation and optimization system based on ACGS algorithm, comprising:

[0008] A spatial light modulator generates a multi-focal plane hologram by modulating the phase and amplitude of multiple image holograms;

[0009] A transflective mirror is arranged in the light propagation area of ​​the multi-focal plane hologram, and the light field of the multi-focal plane hologram is reflected by the transflective mirror and transmitted to the free-form surface mirror;

[0010] The free-form surface mirror is used to reflect and shape the light field of the multi-focal plane hologram, so that the light field of the multi-focal plane hologram is finally reflected into the human eye, and the generated multi-focal plane virtual imaging is used as the target image.

[0011] In addition, the present invention also provides a method for generating and optimizing a multi-focal plane hologram system based on the ACGS algorithm, comprising the following steps:

[0012] Step 100: setting the phase distribution and amplitude of the hologram generated by the spatial light modulator according to the target image information to generate an initial multi-focal plane hologram;

[0013] Step 200, performing a simulation experiment on the initial multi-focal plane hologram, and evaluating the simulation experiment test results through the key indicators of the optical path system and the multi-focal plane hologram to optimize the holographic algorithm of the multi-focal plane hologram and form a multi-focal plane hologram with aberration compensation;

[0014] Step 300: further optimizing the phase distribution of the multi-focal plane hologram by using an iterative algorithm to optimize the virtual imaging quality of the multi-focal plane hologram;

[0015] Step 400: When the number of iterations reaches a preset value, the loop is stopped, and the final phase information and holographic algorithm are introduced into the multi-focal plane hologram.

[0016] As a preferred solution of the present invention, in step 100, a single image hologram is passed through the spatial light modulator to obtain an output image, and the spatial light modulator can quickly process multiple image holograms and obtain multiple output images, and the output images have different focal points, so that all the output images are superimposed together to form the initial multi-focal plane hologram with a three-dimensional display effect;

[0017] The phase distribution of the initial multi-focal plane hologram is randomly generated, and the amplitude of the initial multi-focal plane hologram is set to a fixed value to generate an initialized random phase distribution. As the initial light field, the formula of the initial multi-focal plane hologram is as follows:

[0018]

[0019] According to the phase information of the initial light field, the amplitude of the initial light field is set to a fixed value of 1, and u(x, y) is the initial multi-focal plane hologram in the spatial domain.

[0020] As a preferred solution of the present invention, in step 200, the method for implementing the simulation experiment on the initial multi-focal plane hologram is:

[0021] Converting the initial multi-focal plane hologram into the frequency domain by Fresnel transformation to obtain the complex amplitude distribution of each output image focal plane processed by the spatial light modulator;

[0022] A frequency domain constraint is imposed on each of the output image focal planes, the phase information of each of the output image focal planes is retained, and the calculated frequency domain plane light field amplitude is replaced with a known light field amplitude.

[0023] A dynamic adjustment factor is introduced to combine the image error and the local contrast through a nonlinear combination, and the weight of each output image focal plane is adjusted, and the phase distribution of each output image focal plane is optimized according to the weight.

[0024] As a preferred solution of the present invention, the initial multi-focal plane hologram includes holograms corresponding to different focal planes, and different output image focal planes correspond to different reconstruction depths or different virtual imaging layers, and each output image focal plane will present a different part or layer of the initial multi-focal plane hologram.

[0025] As a preferred solution of the present invention, the initial light field of the initial multi-focal plane hologram is transformed into the spectrum domain through Fresnel transformation to obtain the complex amplitude distribution on the focal plane of j output images.

[0026] Frequency domain constraints are imposed on each output image focal plane, only the phase information of the light field of each output image focal plane in the frequency domain distribution is retained, and each output image focal plane is constrained by a known light field amplitude |F j (u,v)|Replace the calculated frequency domain surface light field amplitude

[0027] The corrected complex amplitude distribution of each output image focal plane Perform inverse Fresnel transform and correct the complex amplitude distribution The light field re-propagates to the input surface of the spatial light modulator

[0028] As a preferred solution of the present invention, the image error between the current image output from each output image focal plane and the ideal image is calculated respectively. With local contrast And introduce nonlinear weight factors and

[0029]

[0030] in represents the light intensity distribution of the ideal image on each output image focal plane, Indicates the calculated light intensity distribution on each output image focal plane. std[] represents the function of calculating the standard deviation, which is used to measure the local contrast. The larger the standard deviation, the higher the local contrast, which means the better the image details and clarity. The smaller the image error, the higher the image quality.

[0031] As a preferred solution of the present invention, a dynamic adjustment factor α is introduced by combining the error weight and the local contrast weight. k , dynamically adjust the weights of the two types of indicators to optimize the contribution of the final hologram, balance the image error and visual contrast, and use exponential nonlinear weighting processing to generate a comprehensive weight

[0032]

[0033] α k =exp(-2k / iterations);

[0034] Iterations is the total number of iterations.

[0035] As a preferred solution of the present invention, in step 300, after each iteration, the phase distributions of the focal planes of the output images are coherently superimposed to obtain multi-focal plane phase distribution information, and the multi-focal plane phase distribution information is used as the input light field for the next iteration until the set number of iterations is reached or the convergence condition is met, wherein the specific implementation method for obtaining the multi-focal plane phase distribution information is:

[0036] The holographic coherence superposition principle is adopted to reconstruct multiple focal planes synchronously, and the error and local contrast between the current image and the ideal image in each iterative reconstruction process are calculated to generate corresponding weights. Among them, the subscripts represent the kth iteration process and the jth focal plane;

[0037] The image error and local contrast are nonlinearly combined, and the dynamic adjustment factor α is used k Adjust the weight ratio of image error and local contrast, and calculate the comprehensive weight of each output image focal plane

[0038] The specific iteration process is as follows:

[0039]

[0040] In the formula, is the phase distribution in the spectrum domain, which represents the phase information of the light field of the initial multi-focal plane hologram in the spectrum domain after each output image focal plane is optimized by the holographic algorithm;

[0041] is the phase distribution in the spatial domain, which represents the phase information of the light field in the spatial domain after the inverse Fresnel transform of each output image focal plane;

[0042] FrT and IFrT represent Fresnel diffraction transform and inverse Fresnel diffraction transform, respectively. (x, y) and (u, v) are the coordinates in the spatial domain and spectral domain, respectively.

[0043] As a preferred embodiment of the present invention, the comprehensive weight Normalize and update the phase distribution of each output image focal plane;

[0044] The phase distribution of multiple focal planes is obtained by coherently superimposing them together;

[0045] Last pair Apply the spatial constraint and use it as the input light field for the k+1th iteration, and continue to perform the above operations;

[0046] When the number of iterations reaches the set number, the loop stops and the phase information is introduced into the hologram.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] The present invention is used for processing synchronous reconstruction of multiple focal plane images. In the multi-focal plane imaging process, phase recovery is achieved by using Fresnel diffraction transform (FrT) and inverse Fresnel transform (IFrT) instead of traditional Fourier transform (FFT). In the iterative process, weights are calculated according to the error and local contrast of each output image focal plane image, and error correction and detail optimization are balanced by dynamically adjusting factors, thereby improving reconstruction speed, reducing information loss and optimizing image quality. Multiple focal plane images are synchronously optimized, avoiding the image overload problem of traditional GS algorithm when processing multiple focal planes. Weight calculation introduces a nonlinear combination of error and local contrast, and error correction and detail optimization are balanced by dynamically adjusting factors, thereby improving image clarity and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0050] Figure 1 A schematic diagram of the overall structure of a multi-focal plane hologram generation and optimization system provided by an embodiment of the present invention;

[0051] Figure 2 A schematic diagram of Fresnel diffraction transformation and inverse Fresnel diffraction transformation provided by an embodiment of the present invention; DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] like Figure 1 As shown, the present invention provides a multi-focal plane hologram generation and optimization system based on ACGS algorithm, including: a spatial light modulator, a transflective mirror and a free-form surface mirror.

[0054] The spatial light modulator forms an output image focal plane by modulating the phase and amplitude of multiple image holograms. All output image focal planes with different focal points are superimposed together to form a multi-focal plane hologram with a three-dimensional display effect.

[0055] The transflective mirror is arranged in the light propagation area of ​​the multi-focal plane hologram, and the light field of the multi-focal plane hologram is reflected by the transflective mirror and transmitted to the free-form surface mirror.

[0056] The free-form surface mirror is used to reflect and shape the light field of the multi-focal plane hologram, so that the light field of the multi-focal plane hologram is finally reflected into the human eye, and the generated multi-focal plane virtual imaging is used as the target image.

[0057] That is to say, this embodiment provides an off-axis catadioptric optical system, which includes an image source, a free-form surface reflector, a flat reflector and a spatial light modulator SLM, wherein the system adopts a single-point diamond processing technology and uses an aluminum-plated film as the material of the reflector. Through ZEMAX simulation calculation, the error of the system sagittal field is 0.3118 mm, the error of the meridian field is 0.0986 mm, the maximum distortion is 0.1380%, the MTF50 value reaches 25lp / mm, and the field of view of the system is 7.5°×3.5°. The system has low aberration and excellent imaging performance, and is suitable for the field of high-precision optical imaging.

[0058] This embodiment uses the GS algorithm to generate the precise phase distribution of the multi-focal plane hologram. Therefore, after the image hologram is loaded into the spatial light modulator SLM, the precise phase distribution is controlled so that the light field of the generated multi-focal plane hologram can accurately control the focal position and depth distribution. Combined with the optical path design of the reflective mirror and the free-form surface mirror, the imaging quality of the multi-focal plane hologram display is ensured, avoiding the limitations of the traditional single-focal plane hologram display.

[0059] In addition, the present invention also provides a method for generating and optimizing a multi-focal plane hologram system based on the ACGS algorithm, comprising the following steps:

[0060] Step 100: Setting the phase distribution and amplitude of the hologram generated by the spatial light modulator according to target image information to generate an initial multi-focal plane hologram.

[0061] Step 200: Perform a simulation experiment on the initial multi-focal plane hologram, and evaluate the simulation experiment test results through the key indicators of the optical path system and the multi-focal plane hologram to optimize the holographic algorithm of the multi-focal plane hologram and form a multi-focal plane hologram with aberration compensation.

[0062] Step 300: further optimize the phase distribution of the multi-focal plane hologram using an iterative algorithm to optimize the virtual imaging quality of the multi-focal plane hologram.

[0063] Step 400: When the number of iterations reaches a preset value, the loop is stopped, and the final phase information and holographic algorithm are introduced into the multi-focal plane hologram.

[0064] This embodiment uses multi-focal plane synchronous reconstruction technology. Compared with the existing two-dimensional display technology, this embodiment can provide users with a real three-dimensional visual effect by reconstructing multi-focal plane light fields. Due to the synchronous reconstruction and optimization of multiple focal plane images, the stereoscopic sense and depth sense of the image are significantly enhanced, allowing users to obtain a stronger immersive experience, overcoming the limitation of traditional two-dimensional display technology that cannot provide depth information.

[0065] In step 100, a single image hologram is passed through the spatial light modulator to obtain an output image, and the spatial light modulator is capable of quickly processing multiple image holograms and obtaining multiple output image focal planes, and the output image focal planes have different focal points, so that all output image focal planes are superimposed together to form the initial multi-focal plane hologram with a three-dimensional display effect.

[0066] The phase distribution of the initial multi-focal plane hologram is randomly generated, and the amplitude of the initial multi-focal plane hologram is set to a fixed value to generate an initial random phase distribution φ random As the initial light field, the formula of the initial multi-focal plane hologram is as follows:

[0067] u(x,y)=exp[iφ random (x,y)];

[0068] According to the phase information of the initial light field, the amplitude of the initial light field is set to a fixed value of 1, and u(x, y) is the initial multi-focal plane hologram in the spatial domain.

[0069] In step 200, the method for implementing the simulation experiment on the initial multi-focal plane hologram is:

[0070] The initial multi-focal plane hologram is converted into the frequency domain by Fresnel transformation to obtain the complex amplitude distribution of each output image focal plane processed by the spatial light modulator, wherein the initial multi-focal plane hologram includes holograms corresponding to different focal planes, and different output image focal planes correspond to different reconstruction depths or different virtual imaging layers, and each output image focal plane will present a different part or layer of the initial multi-focal plane hologram.

[0071] A frequency domain constraint is imposed on each of the output image focal planes, the phase information of each of the output image focal planes is retained, and the calculated frequency domain plane light field amplitude is replaced with a known light field amplitude.

[0072] A dynamic adjustment factor is introduced to combine the image error and the local contrast through a nonlinear combination, and the weight of each output image focal plane is adjusted, and the phase distribution of each output image focal plane is optimized according to the weight.

[0073] This process is a part of the simulation experiment on the initial multi-focal plane hologram, and is mainly to further optimize the light field of the multi-focal plane hologram.

[0074] During this process, the system optimizes the phase distribution on the focal plane of each output image through dynamic adjustment factors and iterative algorithms based on specific optimization goals.

[0075] Image error generally refers to the difference between the target image and the actual reconstructed image in the initial multi-focal plane hologram reconstruction process. In this embodiment, the image error is the difference between the light intensity or phase distribution of the target image and the reconstructed image on each output image focal plane.

[0076] Local contrast refers to the degree of light intensity variation in a small area during each reconstruction of the initial multi-focal hologram, reflecting the clarity of the details in the area. Areas with higher contrast usually have more obvious details and edges.

[0077] The object of local contrast is the light intensity distribution of the local area of ​​the image. During the simulation process, the system calculates the contrast of the local area in the target image and uses the contrast to evaluate the imaging quality on the focal plane of each output image.

[0078] Furthermore, in this embodiment, the key indicators of the optical path system and the multi-focal plane hologram are specifically: imaging clarity, focal length accuracy and aberration correction effect.

[0079] Image error usually refers to the difference between an ideal image and an actually generated image. The smaller the error, the closer the image reconstruction effect is to the target image and the higher the imaging clarity. Therefore, the reduction of image error directly improves the image clarity. Therefore, the imaging clarity and image error are inversely related. The smaller the image error, the higher the imaging clarity.

[0080] Local contrast is an indicator to measure the clarity of details and edges in an image, and is usually evaluated by calculating the light intensity change (standard deviation) in a local area of ​​the image. The higher the local contrast, the clearer the details and edges of the image, and the better the imaging quality. Therefore, local contrast is positively correlated with imaging clarity. The higher the local contrast, the higher the imaging clarity.

[0081] Focal length accuracy refers to whether the focal lengths of each focal plane image in the optical path system are accurately aligned. If the focal length accuracy is high, the multi-focal plane system can correctly reconstruct images on different focal planes, avoiding image blur or focus confusion, thereby reducing image errors. Therefore, the higher the focal length accuracy, the smaller the image error. If the focal length is inaccurate, it will cause the image to be out of focus or poorly reconstructed on multiple focal planes, thereby increasing image errors.

[0082] The goal of aberration correction is to eliminate or reduce image distortion caused by defects in the optical system through optical design (such as optimization of free-form mirrors). Good aberration correction means that the propagation of the light field will not be affected by unnecessary distortion or deviation, so that the target image can be better restored.

[0083] In summary: image error is negatively correlated with image clarity. The smaller the error, the higher the clarity.

[0084] The local contrast is positively correlated with the image clarity. The higher the local contrast, the better the clarity.

[0085] The focal length accuracy is negatively correlated with the image error. The higher the focal length accuracy, the smaller the image error.

[0086] The aberration correction effect is negatively correlated with the image error and local contrast. The better the aberration correction effect, the smaller the image error and the higher the local contrast, thereby improving the imaging quality.

[0087] Therefore, based on the image error and local contrast of each output image focal plane corresponding to each simulation, the imaging clarity, focal length accuracy, and aberration correction effect corresponding to the output image focal plane are determined. Based on the imaging clarity, focal length accuracy, and aberration correction effect corresponding to the output image focal plane, the phase distribution on each output image focal plane is optimized through dynamic adjustment factors and iterative algorithms.

[0088] That is Figure 2 As shown in the figure, the initial light field of the initial multi-focal plane hologram is transformed into the spectrum domain through Fresnel transformation, and the complex amplitude distribution on the focal plane of the j output images is obtained.

[0089] Frequency domain constraints are imposed on each output image focal plane, only the phase information of the light field of each output image focal plane in the frequency domain distribution is retained, and each output image focal plane is constrained by a known light field amplitude |F j (u,v)|Replace the calculated frequency domain surface light field amplitude

[0090] The corrected complex amplitude distribution of each output image focal plane Perform inverse Fresnel transform and correct the complex amplitude distribution The light field re-propagates to the input surface of the spatial light modulator

[0091] This implementation optimizes the quality of multi-focal plane images by combining multi-focal plane synchronous reconstruction technology with dynamic weight adjustment of image error and local contrast. Compared with existing two-dimensional display technology, this application can provide users with a true three-dimensional visual effect through the reconstruction of multi-focal plane light fields. Due to the synchronous reconstruction and optimization of multiple focal plane images, the stereoscopic sense and depth sense of the image are significantly enhanced, allowing users to obtain a stronger immersive experience, overcoming the limitation of traditional two-dimensional display technology that cannot provide depth information.

[0092] In addition, this embodiment adopts Fresnel diffraction transform (FrT) and inverse Fresnel diffraction transform (IFrT) instead of traditional Fourier transform (FFT) to achieve phase recovery of multi-focal plane imaging. Fresnel diffraction transform can effectively reduce the computational complexity when reconstructing multiple image planes and optimize the phase distribution of each output image focal plane image.

[0093] In addition, since the weights are adjusted dynamically in each iteration, the system can maintain a high degree of stability, avoiding the problem of the traditional GS algorithm easily falling into the local optimal solution, and ensuring the accuracy and efficiency of multi-focal plane image reconstruction.

[0094] Among them, the image error between the current image output from each output image focal plane and the ideal image is calculated separately With local contrast And introduce nonlinear weight factors and

[0095]

[0096] in represents the light intensity distribution of the ideal image on each output image focal plane, Indicates the calculated light intensity distribution on each output image focal plane. std[] represents the function of calculating the standard deviation, which is used to measure the local contrast. The larger the standard deviation, the higher the local contrast, which means the better the image details and clarity. The smaller the image error, the higher the image quality.

[0097] Combine the error weight and the local contrast weight to introduce a dynamic adjustment factor α k , dynamically adjust the weights of the two types of indicators to optimize the contribution of the final hologram, balance the image error and visual contrast, and use exponential nonlinear weighting processing to generate a comprehensive weight

[0098]

[0099] α k =exp(-2k / iterations);

[0100] Iterations is the total number of iterations.

[0101] This implementation optimizes the phase distribution of the image by dynamically adjusting the weights of image error and local contrast. Since the optimized optical elements (such as free-form surface reflectors) can better control the propagation path of light, the spherical aberration, astigmatism and other distortions in the optical system are reduced, thereby effectively improving the clarity and authenticity of the imaging. The reduction in image distortion greatly improves the quality of the final displayed image, especially in multi-focal plane imaging scenarios, which can significantly reduce information loss and image distortion.

[0102] This implementation is optimized on the basis of the traditional GS algorithm, and adopts the principle of holographic coherence superposition to reconstruct multiple focal planes synchronously. For each iterative reconstruction process, the corresponding weight is calculated based on the error and local contrast between the current image and the ideal image. The subscripts represent the kth iteration process, the jth focal plane, and the nonlinear combination of image error and local contrast. Then, according to the dynamic adjustment factor α k Adjust the ratio of image error and local contrast weight to calculate the comprehensive weight of each output image focal plane Dynamically adjust the weights of the two types of indicators to optimize the contribution of the final hologram to the final hologram, balancing image error and visual contrast.

[0103] And the dynamic adjustment factor α k , which decays smoothly from an initial value close to 1 to close to 0 during the entire iteration process. In the first half of the iteration, the image error weight is used as the main evaluation index, and the local contrast weight has little effect on the optimization target phase. In the second half of the iteration, as α k The decrease of local contrast weight is more emphasized as the main evaluation index, supplemented by image error weight, to optimize the target phase. Here, the numerator coefficient 2 is the transition coefficient, which can effectively adjust the attenuation rate of the dynamic adjustment factor to achieve a balance between error correction and detail optimization.

[0104] The speed of image reconstruction is significantly improved by optimizing the weight ratio of error and local contrast through dynamic adjustment factors. In addition, through fine weight allocation, information loss in the image reconstruction process is reduced, the speed and quality of image reconstruction are guaranteed, the local optimal solution problem in traditional methods is avoided, and the efficiency of the entire reconstruction process is improved.

[0105] This implementation combines the GS algorithm with the spatial light modulator SLM, which directly brings about high-precision dynamic modulation of the phase distribution, realizes multi-focal plane display of the same image content, and improves the applicability and display effect of the system: based on Matlab simulation design and iterative optimization of the GS algorithm, it ensures that the phase distribution generated by the hologram is highly consistent with the target image, accurately grasps the focal length, and achieves a three-dimensional imaging effect.

[0106] This implementation proposes a nonlinear weight adjustment mechanism based on image error and local contrast, and balances the relationship between error correction and detail optimization by introducing a dynamic adjustment factor. This dynamic adjustment mechanism significantly improves the quality of the image. In the early stages of the algorithm, it mainly relies on image error correction, and in the later stages, as the error weight gradually decreases, more attention is paid to improving local contrast, thereby ensuring that the details of the image are clearer, and the realism and detail expression of the image are enhanced. This optimization can be dynamically adjusted according to the needs of the image at different stages, thereby improving the visual quality of the image.

[0107] In step 300, after each iteration, the phase distributions of the focal planes of the output images are coherently superimposed to obtain multi-focal plane phase distribution information, which is used as the input light field for the next iteration until the set number of iterations is reached or the convergence condition is met. The specific implementation method for obtaining the multi-focal plane phase distribution information is as follows:

[0108] The holographic coherence superposition principle is adopted to reconstruct multiple focal planes synchronously, and the error and local contrast between the current image and the ideal image in each iterative reconstruction process are calculated to generate corresponding weights. Among them, the subscripts represent the kth iteration process and the jth focal plane;

[0109] The image error and local contrast are nonlinearly combined, and the dynamic adjustment factor α is used k Adjust the weight ratio of image error and local contrast, and calculate the comprehensive weight of each output image focal plane

[0110] The specific iteration process is as follows:

[0111]

[0112] In the formula, is the phase distribution in the spectrum domain, which represents the phase information of the light field of the initial multi-focal plane hologram in the spectrum domain after each output image focal plane is optimized by the holographic algorithm;

[0113] is the phase distribution in the spatial domain, which represents the phase information of the light field in the spatial domain after the inverse Fresnel transform of each output image focal plane;

[0114] FrT and IFrT represent Fresnel diffraction transform and inverse Fresnel diffraction transform, respectively. (x, y) and (u, v) are the coordinates in the spatial domain and spectral domain, respectively.

[0115] The comprehensive weight Normalize and update the phase distribution of each output image focal plane;

[0116] The phase distribution of multiple focal planes is obtained by coherently superimposing them together;

[0117] Last pair Apply the spatial constraint and use it as the input light field for the k+1th iteration, and continue to perform the above operations;

[0118] When the number of iterations reaches the set number, the loop stops and the phase information is introduced into the hologram.

[0119] The phase pattern loaded by the spatial light modulator SLM can be adjusted in real time, allowing the position and distribution of the focal plane of the light field to change dynamically and achieve a millisecond-level response speed. This technical feature ensures the stability of the system's display effect in dynamic scenes and can adapt to the needs of complex application scenarios, such as interactive virtual display or real-time augmented reality applications.

[0120] This embodiment is used to process the synchronous reconstruction of multiple focal plane images. In the multi-focal plane imaging process, phase recovery is achieved by using Fresnel diffraction transform (FrT) and inverse Fresnel transform (IFrT) instead of traditional Fourier transform (FFT). In the iterative process, weights are calculated according to the error and local contrast of each output image focal plane image, and error correction and detail optimization are balanced by dynamically adjusting factors, thereby improving reconstruction speed, reducing information loss and optimizing image quality. Multiple focal plane images are optimized synchronously, avoiding the image overload problem of the traditional GS algorithm when processing multiple focal planes. The weight calculation introduces a nonlinear combination of error and local contrast, and the error correction and detail optimization are balanced by dynamically adjusting factors to improve image clarity and quality.

[0121] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A multi-focal plane hologram generation and optimization system based on ACGS algorithm, characterized in that: include: A spatial light modulator generates a multi-focal plane hologram by modulating the phase and amplitude of multiple image holograms; A transflective mirror is arranged in the light propagation area of ​​the multi-focal plane hologram, and the light field of the multi-focal plane hologram is reflected by the transflective mirror and transmitted to the free-form surface mirror; The free-form surface mirror is used to reflect and shape the light field of the multi-focal plane hologram, so that the light field of the multi-focal plane hologram is finally reflected into the human eye, and the generated multi-focal plane virtual imaging is used as the target image.

2. The method of a multi-focal plane hologram generation and optimization system based on ACGS algorithm according to claim 1, characterized in that: The following steps are involved: Step 100: setting the phase distribution and amplitude of the hologram generated by the spatial light modulator according to the target image information to generate an initial multi-focal plane hologram; Step 200, performing a simulation experiment on the initial multi-focal plane hologram, and evaluating the simulation experiment test results through the key indicators of the optical path system and the multi-focal plane hologram to optimize the holographic algorithm of the multi-focal plane hologram and form a multi-focal plane hologram with aberration compensation; Step 300: further optimizing the phase distribution of the multi-focal plane hologram by using an iterative algorithm to optimize the virtual imaging quality of the multi-focal plane hologram; Step 400: When the number of iterations reaches a preset value, the loop is stopped, and the final phase information and holographic algorithm are introduced into the multi-focal plane hologram.

3. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 2, characterized in that: In step 100, a single image hologram is passed through the spatial light modulator to obtain an output image focal plane, and the spatial light modulator is capable of quickly processing multiple image holograms and obtaining multiple output image focal planes, the output image focal planes having different focal points, so that all the output image focal planes are superimposed together to form the initial multi-focal plane hologram with a three-dimensional display effect; The phase distribution of the initial multi-focal plane hologram is randomly generated, and the amplitude of the initial multi-focal plane hologram is set to a fixed value to generate an initialized random phase distribution. As the initial light field, the formula of the initial multi-focal plane hologram is as follows: According to the phase information of the initial light field, the amplitude of the initial light field is set to a fixed value of 1, and u(x, y) is the initial multi-focal plane hologram in the spatial domain.

4. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 2, characterized in that: In step 200, the method for implementing the simulation experiment on the initial multi-focal plane hologram is: Converting the initial multi-focal plane hologram into the frequency domain by Fresnel transformation to obtain the complex amplitude distribution of each output image focal plane processed by the spatial light modulator; A frequency domain constraint is imposed on each of the output image focal planes, the phase information of each of the output image focal planes is retained, and the calculated frequency domain plane light field amplitude is replaced with a known light field amplitude. A dynamic adjustment factor is introduced to combine the image error and the local contrast through a nonlinear combination, and the weight of each output image focal plane is adjusted, and the phase distribution of each output image focal plane is optimized according to the weight.

5. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 4, characterized in that: The initial multi-focal plane hologram includes holograms corresponding to different focal planes, and different output image focal planes correspond to different reconstruction depths or different virtual imaging layers, and each output image focal plane presents a different part or layer of the initial multi-focal plane hologram.

6. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 4, characterized in that: The initial light field of the initial multi-focal plane hologram is transformed into the spectrum domain through Fresnel transformation to obtain the complex amplitude distribution on the focal plane of j output images. Frequency domain constraints are imposed on each output image focal plane, only the phase information of the light field of each output image focal plane in the frequency domain distribution is retained, and each output image focal plane is constrained by a known light field amplitude |F j (u,v)|Replace the calculated frequency domain surface light field amplitude The corrected complex amplitude distribution of each output image focal plane Complex amplitude distribution after inverse Fresnel transformation and correction The light field re-propagates to the input surface of the spatial light modulator 7. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 6, characterized in that: Calculate the image error between the current image output from each output image focal plane and the ideal image in represents the light intensity distribution of the ideal image on each output image focal plane, Indicates the calculated light intensity distribution on each output image focal plane. std[] represents the function of calculating the standard deviation, which is used to measure the local contrast. The larger the standard deviation, the higher the local contrast, which means the better the image details and clarity. The smaller the image error, the higher the image quality.

8. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 7, characterized in that: Combine the error weight and the local contrast weight to introduce a dynamic adjustment factor α k , dynamically adjust the weights of the two types of indicators to optimize the contribution of the final hologram, balance the image error and visual contrast, and use exponential nonlinear weighting processing to generate a comprehensive weight α k =exp(-2k / iterations); Iterations is the total number of iterations.

9. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 8, characterized in that: In step 300, after each iteration, the phase distributions of the focal planes of the output images are coherently superimposed to obtain multi-focal plane phase distribution information, which is used as the input light field for the next iteration until the set number of iterations is reached or the convergence condition is met. The specific implementation method for obtaining the multi-focal plane phase distribution information is as follows: The holographic coherence superposition principle is adopted to reconstruct multiple focal planes synchronously, and the error and local contrast between the current image and the ideal image in each iterative reconstruction process are calculated to generate corresponding weights. Among them, the subscripts represent the kth iteration process and the jth focal plane; The image error and local contrast are nonlinearly combined, and the dynamic adjustment factor α is used k Adjust the weight ratio of image error and local contrast, and calculate the comprehensive weight of each output image focal plane The specific iteration process is as follows: In the formula, is the phase distribution in the spectrum domain, which represents the phase information of the light field of the initial multi-focal plane hologram in the spectrum domain after each output image focal plane is optimized by the holographic algorithm; is the phase distribution in the spatial domain, which represents the phase information of the light field in the spatial domain after the inverse Fresnel transform of each output image focal plane; FrT and IFrT represent Fresnel diffraction transform and inverse Fresnel diffraction transform, respectively. (x, y) and (u, v) are the coordinates in the spatial domain and spectral domain, respectively.

10. The method for generating and optimizing a multi-focal plane hologram based on the ACGS algorithm according to claim 9, characterized in that: The comprehensive weight Normalize and update the phase distribution of each output image focal plane; The phase distribution of multiple focal planes is obtained by coherently superimposing them together; Last pair Apply the spatial constraint and use it as the input light field for the k+1th iteration, and continue to perform the above operations; When the number of iterations reaches the set number, the loop stops and the phase information is introduced into the hologram.

Images